INDEX

Explanations

highly active verbs and nouns that indicate significant actions or elements within a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

bourne

-0.15

insky

-0.15

ayer

-0.15

ActionCreators

-0.14

endi

-0.14

bs

-0.14

aturas

-0.14

borne

-0.13

 sortable

-0.13

Î²Î¬

-0.13

POSITIVE LOGITS

 manip

0.15

ÙĬÙĨØ©

0.15

enny

0.14

unner

0.14

ISO

0.14

 achie

0.14

owler

0.14

anik

0.14

ä¸ĸ

0.14

uja

0.14

Activations Density 0.002%

highly active verbs and nouns that indicate significant actions or elements within a text

No Comments

No Known Activations