INDEX

Explanations

common words and phrases related to actions or states occurring within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

wat

-0.17

ritel

-0.15

Sad

-0.15

 perms

-0.15

gende

-0.15

inkle

-0.15

Gul

-0.14

LEN

-0.14

wr

-0.14

ael

-0.14

POSITIVE LOGITS

trak

0.17

.Formatter

0.15

baugh

0.15

uos

0.14

Neo

0.14

è¿¹

0.14

Å©

0.14

pty

0.14

volution

0.14

 Skinny

0.14

Activations Density 0.007%

common words and phrases related to actions or states occurring within various contexts

No Comments

No Known Activations