INDEX

Explanations

phrases that denote relationships or associations, particularly involving the word "of."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

The neuron activates for the phrase "one of" when it appears in contexts describing membership, affiliation, or being part of a group or collection.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

constructions that identify someone or something as a single member of a specified group or subset (the “one of …” partitive pattern).

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.scalablytyped

-0.17

anker

-0.16

obia

-0.16

 paramName

-0.15

otine

-0.15

"]->

-0.14

 Ging

-0.13

.Strings

-0.13

iche

-0.13

POSITIVE LOGITS

few

0.20

 many

0.16

Few

0.16

strup

0.15

 several

0.15

.daily

0.15

_many

0.14

ÅĽ

0.14

many

0.14

inde

0.14

Activations Density 0.063%

phrases that denote relationships or associations, particularly involving the word "of."

The neuron activates for the phrase "one of" when it appears in contexts describing membership, affiliation, or being part of a group or collection.

constructions that identify someone or something as a single member of a specified group or subset (the “one of …” partitive pattern).

No Comments

No Known Activations