INDEX

Explanations

phrases and structures indicating membership or affiliation with specific groups or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

Ã©gor

-0.18

à¸²à¸ĩ

-0.17

DisplayStyle

-0.16

çĮª

-0.15

_ASSUME

-0.14

vier

-0.14

 ÐłÐµÐ³

-0.14

zer

-0.14

 Ð¿Ð¾Ð±Ð°

-0.14

 nhau

-0.14

POSITIVE LOGITS

 whom

0.19

bble

0.15

0.14

 late

0.14

chan

0.13

Ä±ÄŁÄ±

0.13

ãĥ£

0.13

rowse

0.13

 Fridays

0.13

Activations Density 0.033%

phrases and structures indicating membership or affiliation with specific groups or organizations

No Comments

No Known Activations