INDEX

Explanations

the word "subject" and related terms or references to subjects in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ogue

-0.16

ereum

-0.16

Enumerator

-0.16

eful

-0.15

unta

-0.14

ÙħØª

-0.14

è²´

-0.14

PF

-0.14

 Meadows

-0.14

POSITIVE LOGITS

æīķ

0.17

ideo

0.16

IVEN

0.15

ìŀĲìĿ¸

0.15

icc

0.15

iec

0.14

acute

0.14

idy

0.13

stell

0.13

iyon

0.13

Activations Density 0.012%

the word "subject" and related terms or references to subjects in discussions

No Comments

No Known Activations