INDEX

Explanations

phrases related to perception or interpretation of the world

oai_token-act-pair · gpt-3.5-turbo

expressions of perception and personal beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

auga

-0.77

ARS

-0.66

 entirety

-0.62

txt

-0.61

amiya

-0.61

fts

-0.60

icken

-0.60

umbnail

-0.59

gart

-0.58

ousing

-0.57

POSITIVE LOGITS

 interacting

0.77

 interacts

0.74

uate

0.71

 interactions

0.70

 communicating

0.69

 differs

0.68

 things

0.66

 versus

0.66

 interact

0.65

 interpreting

0.65

Activations Density 0.213%

phrases related to perception or interpretation of the world

expressions of perception and personal beliefs

No Comments

No Known Activations