INDEX

Explanations

mentions of explosive devices and bomb-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

isko

-0.17

 Backbone

-0.16

razier

-0.15

 irrational

-0.15

 microscope

-0.14

vecs

-0.14

ìĨĲ

-0.13

ãĤ´ãĥª

-0.13

POSITIVE LOGITS

 deton

0.45

 explosive

0.43

 explosion

0.40

 Explos

0.38

çĪĨ

0.38

 explosives

0.38

 explode

0.37

 explodes

0.36

çĤ¸

0.35

 explosions

0.35

Activations Density 0.103%

mentions of explosive devices and bomb-related terms

No Comments

No Known Activations