INDEX

Explanations

references to legal or authoritative sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

å¾Ĳ

-0.16

ONTAL

-0.16

uar

-0.15

ìļ°

-0.15

resse

-0.14

unks

-0.14

gles

-0.14

idor

-0.14

toc

-0.14

anni

-0.14

POSITIVE LOGITS

 rall

0.15

Mus

0.15

PathComponent

0.14

è°±

0.14

 Lance

0.13

 Vintage

0.13

hap

0.13

Ð»Ð°Ð½

0.13

eld

0.13

assen

0.13

Activations Density 0.002%

references to legal or authoritative sources

No Comments

No Known Activations