INDEX

Explanations

references to historical violence and consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

gression

-0.17

 Hurt

-0.14

":[{↵

-0.14

Ú©Ø§Ø±ÛĮ

-0.14

 Assault

-0.13

mpeg

-0.13

iji

-0.13

·»

-0.13

AP

-0.13

rips

-0.13

POSITIVE LOGITS

 execution

0.48

 hang

0.47

 hanging

0.42

hang

0.41

Hang

0.41

 executions

0.41

 Hang

0.40

 executed

0.40

 Execution

0.39

 Hanging

0.38

Activations Density 0.202%

references to historical violence and consequences

No Comments

No Known Activations