INDEX

Explanations

quantitative measures and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

etty

-0.15

heel

-0.14

edelta

-0.14

ilon

-0.14

 fewer

-0.14

Few

-0.13

HEEL

-0.13

 Feinstein

-0.13

uali

-0.13

_HS

-0.13

POSITIVE LOGITS

 fold

0.48

 times

0.47

-fold

0.45

times

0.44

 TIMES

0.43

åĢį

0.42

fold

0.41

-times

0.40

 folds

0.40

 Fold

0.39

Activations Density 0.080%

quantitative measures and comparisons

No Comments

No Known Activations