INDEX

Explanations

quantities and relationships within complex scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

sembl

-0.15

stÃ¥

-0.14

iev

-0.14

udem

-0.13

rawn

-0.13

APER

-0.12

 Bison

-0.12

illo

-0.12

ĶåĽŀ

-0.12

owski

-0.12

POSITIVE LOGITS

/*č↵

0.15

acÃŃ

0.15

//{{

0.14

ÐµÑĢÐ¾

0.14

afort

0.13

Disallow

0.13

ë¶Ħ

0.13

junction

0.13

#ac

0.13

rut

0.13

Activations Density 0.779%

quantities and relationships within complex scenarios

No Comments

No Known Activations