INDEX

Explanations

specific patterns of characters or symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

argo

-0.17

asn

-0.16

urrent

-0.14

_RELEASE

-0.14

iÐ²

-0.14

 Current

-0.13

çīĮ

-0.13

mark

-0.13

Compilation

-0.13

noch

-0.13

POSITIVE LOGITS

 themselves

0.40

 itself

0.36

 himself

0.36

 herself

0.35

 Himself

0.32

 yourself

0.31

 oneself

0.31

 ourselves

0.30

 myself

0.29

 yourselves

0.29

Activations Density 0.009%

specific patterns of characters or symbols

No Comments

No Known Activations