INDEX

Explanations

non-English characters or symbols

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

 ÐĴÐ¸ÐºÐ¸

-0.17

zell

-0.15

é¾Ħ

-0.15

Ann

-0.14

çĴ°

-0.14

ì¡´

-0.14

 boca

-0.14

enia

-0.14

Ã²i

-0.14

nu

-0.13

POSITIVE LOGITS

 Rose

0.16

Por

0.15

por

0.15

 rose

0.15

 original

0.15

rose

0.15

 behind

0.15

 Rosa

0.15

 current

0.14

 relief

0.14

Activations Density 0.007%

non-English characters or symbols

No Comments

No Known Activations