INDEX

Explanations

references to the source or origin of items or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

å¯Ŀ

-0.16

anke

-0.15

ç¿¼

-0.15

jang

-0.15

rollo

-0.14

/rfc

-0.14

PermissionsResult

-0.14

áº¿

-0.13

 belir

-0.13

rotch

-0.13

POSITIVE LOGITS

ucker

0.16

ubar

0.16

ERC

0.15

Dip

0.15

/to

0.15

 Jewel

0.15

hue

0.14

CodeGen

0.14

mine

0.14

oth

0.13

Activations Density 0.149%

references to the source or origin of items or actions

No Comments

No Known Activations