INDEX

Explanations

phrases that emphasize beliefs or assertions in a discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ascade

-0.16

 intptr

-0.15

ØªØ¨Ùĩ

-0.14

udur

-0.14

igor

-0.14

anton

-0.13

ê¼

-0.13

ngrx

-0.13

ÑĩÐµ

-0.13

EdgeInsets

-0.13

POSITIVE LOGITS

why

0.18

why

0.16

etti

0.15

 roles

0.14

Bak

0.14

alli

0.14

roles

0.14

 partially

0.13

-Ta

0.13

 especially

0.13

Activations Density 0.051%

phrases that emphasize beliefs or assertions in a discussion

No Comments

No Known Activations