INDEX

Explanations

words and word fragments that start with "im" or "im" occurring in words

oai_token-act-pair · gpt-3.5-turbo

prefixes indicating immaturity or inadequacy

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

 lodge

-0.73

ttes

-0.70

 tipped

-0.63

Rav

-0.62

 butterfly

-0.62

IGH

-0.61

 underside

-0.60

 Cullen

-0.59

 cake

-0.58

 independently

-0.57

POSITIVE LOGITS

itating

1.37

itations

1.32

balanced

1.28

itates

1.21

itated

1.18

itative

1.12

manent

1.12

mer

1.12

bal

1.08

mediate

1.04

Activations Density 0.009%

words and word fragments that start with "im" or "im" occurring in words

prefixes indicating immaturity or inadequacy

No Comments

No Known Activations