INDEX

Explanations

expressions of agreement or justification in conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

uy

-0.15

Äįan

-0.14

uri

-0.14

lash

-0.14

ingham

-0.14

tons

-0.14

åĳ¢

-0.13

 credits

-0.13

uya

-0.13

-spin

-0.13

POSITIVE LOGITS

cha

0.16

Cha

0.15

 nature

0.15

ouser

0.15

cha

0.14

 Nature

0.13

Ŀi

0.13

 Provid

0.13

nature

0.13

Ä±rak

0.13

Activations Density 0.028%

expressions of agreement or justification in conversation

No Comments

No Known Activations