INDEX

Explanations

qualities and characteristics related to items or concepts that denote value and complexity

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

lessness

-0.21

 rencont

-0.17

vr

-0.15

oms

-0.14

naments

-0.14

imento

-0.14

precation

-0.14

orgia

-0.14

Giz

-0.14

ä»¬

-0.14

POSITIVE LOGITS

 enough

0.35

ly

0.28

 Enough

0.23

LY

0.19

ised

0.17

ä¸Ķ

0.17

 affairs

0.17

mente

0.17

AF

0.17

ized

0.17

Activations Density 0.811%

qualities and characteristics related to items or concepts that denote value and complexity

No Known Activations