INDEX

Explanations

URLs or links in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

elman

-0.19

aternity

-0.16

ipl

-0.15

ãģ£ãģį

-0.15

Ã»t

-0.14

á»Ľi

-0.14

itel

-0.14

idelberg

-0.14

renc

-0.14

zano

-0.14

POSITIVE LOGITS

forest

0.15

Rug

0.15

squ

0.15

CAP

0.14

 credit

0.14

 Cooper

0.14

atra

0.14

Lid

0.14

squ

0.13

Ã³m

0.13

Activations Density 0.021%

URLs or links in the text

No Comments

No Known Activations