INDEX

Explanations

negations or words expressing denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ally

-0.15

ÐµÐ³Ð°

-0.15

 å¡

-0.14

uitka

-0.14

ActionCreators

-0.14

ä¸įåĪ°

-0.14

eyen

-0.13

 McGr

-0.13

DownList

-0.13

POSITIVE LOGITS

ori

0.18

 necessarily

0.17

ches

0.15

eworthy

0.15

 bent

0.15

axon

0.15

ÑĨÐ¸

0.15

ched

0.15

aken

0.14

zsche

0.14

Activations Density 0.050%

negations or words expressing denial

No Comments

No Known Activations