INDEX

Explanations

mentions of groupings or comparisons between entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ILog

-0.15

adem

-0.15

aso

-0.14

tol

-0.14

adia

-0.14

Ð¸Ð½Ñĸ

-0.14

addtogroup

-0.13

/pm

-0.13

agy

-0.13

Ð°

-0.13

POSITIVE LOGITS

aver

0.15

 Ches

0.14

NGX

0.14

setattr

0.14

Dw

0.13

odox

0.13

FFFFFFFF

0.13

ÌĨ

0.13

osg

0.13

ufs

0.13

Activations Density 0.321%

mentions of groupings or comparisons between entities

No Comments

No Known Activations