INDEX

Explanations

references to aggressive behavior or concepts related to aggression

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

eder

-0.19

ario

-0.15

à¤¾à¤ī

-0.15

umb

-0.14

rhs

-0.14

ucky

-0.14

uki

-0.14

anel

-0.14

annah

-0.14

ahu

-0.14

POSITIVE LOGITS

 Throne

0.15

PRS

0.15

ymax

0.15

ç¿¼

0.14

ulos

0.14

 compar

0.14

-factor

0.14

 definite

0.13

Boo

0.13

angement

0.13

Activations Density 0.015%

references to aggressive behavior or concepts related to aggression

No Comments

No Known Activations