INDEX

Explanations

characters or symbols commonly associated with digital encoding or textual representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Masc

-0.15

ouro

-0.15

 tween

-0.15

redits

-0.14

ww

-0.14

orda

-0.14

otland

-0.13

vertime

-0.13

omez

-0.13

wich

-0.13

POSITIVE LOGITS

Sad

0.17

yog

0.17

Sad

0.16

 Dimensions

0.15

 yoga

0.15

kro

0.15

Cro

0.15

 spirituality

0.14

sad

0.14

 spiritual

0.14

Activations Density 0.005%

characters or symbols commonly associated with digital encoding or textual representation

No Comments

No Known Activations