INDEX

Explanations

references to literary works and authors

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

é§

-0.16

uelles

-0.15

ŀĭ

-0.14

pv

-0.14

ragment

-0.14

ãģĹãĤĩãģĨ

-0.14

azu

-0.14

edic

-0.14

èĢĲ

-0.14

 bree

-0.13

POSITIVE LOGITS

Jew

0.21

Wor

0.20

utsch

0.20

Paw

0.20

 Chow

0.18

ije

0.18

isch

0.17

Saw

0.17

Ds

0.17

Aw

0.17

Activations Density 0.031%

references to literary works and authors

No Comments

No Known Activations