INDEX

Explanations

references to geographic locations, cultural groups, and organizations

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

Ø¯Ø§ÙĨ

-0.17

oton

-0.17

sdale

-0.15

Ð¶Ñĥ

-0.15

earing

-0.14

manÄ±

-0.14

Uri

-0.14

 outs

-0.14

allet

-0.14

cdn

-0.14

POSITIVE LOGITS

Ã

0.15

 éĵ

0.14

-License

0.14

owntown

0.14

thalm

0.14

åĿª

0.13

QUENCE

0.13

Âĥ

0.13

DonaldTrump

0.13

à¹Ħà¸¥

0.13

Activations Density 0.236%

references to geographic locations, cultural groups, and organizations

No Comments

No Known Activations