INDEX

Explanations

instances of the word "under" and its variations, suggesting a focus on hierarchical or structural relationships

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

The neuron activates for the Swedish/German word "under" (meaning "sub" or "under"), particularly in taxonomic or classification contexts where it precedes "arter" (species/subspecies).

oai_token-act-pair · claude-4-5-sonnet 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

sett

-0.18

ihar

-0.17

 Berger

-0.15

alaria

-0.15

rica

-0.15

á»ĵng

-0.15

ptom

-0.14

ulin

-0.14

illis

-0.14

obot

-0.14

POSITIVE LOGITS

iect

0.16

 rele

0.14

EO

0.14

urst

0.14

_HTML

0.14

ender

0.14

DK

0.13

.arr

0.13

nd

0.13

Activations Density 0.004%

instances of the word "under" and its variations, suggesting a focus on hierarchical or structural relationships

The neuron activates for the Swedish/German word "under" (meaning "sub" or "under"), particularly in taxonomic or classification contexts where it precedes "arter" (species/subspecies).

No Comments

No Known Activations