INDEX

Explanations

references to specific cultural or entertainment topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

lesc

-0.16

)did

-0.15

[__

-0.15

 carrier

-0.15

carrier

-0.15

Ð»ÐµÑĩ

-0.14

ì£¼ìĭľ

-0.14

ÏĥÎ¼Î±ÏĦÎ±

-0.14

æ´²

-0.14

lec

-0.14

POSITIVE LOGITS

 matters

0.19

 stories

0.19

 topics

0.19

 breaking

0.19

 beat

0.18

 issues

0.18

 story

0.17

-breaking

0.16

 subjects

0.16

 developments

0.16

Activations Density 0.034%

references to specific cultural or entertainment topics

No Comments

No Known Activations