INDEX

Explanations

references to licenses and legal information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ons

-0.16

cripts

-0.15

oner

-0.15

rug

-0.14

leans

-0.14

ands

-0.13

asters

-0.13

Nach

-0.13

ittel

-0.13

ount

-0.13

POSITIVE LOGITS

aleza

0.15

åĿ

0.15

Sax

0.14

nia

0.14

blas

0.14

Hook

0.14

kÃ¼

0.14

utut

0.14

ewis

0.13

||(

0.13

Activations Density 0.012%

references to licenses and legal information

No Comments

No Known Activations