INDEX

Explanations

phrases related to criticism or negative evaluations

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the concept of beginnings or starting points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ogie

-0.82

ormal

-0.66

plot

-0.66

had

-0.64

reau

-0.64

xual

-0.64

artment

-0.64

share

-0.63

ildo

-0.62

okes

-0.62

POSITIVE LOGITS

 standpoint

1.71

 outset

1.60

 perspective

1.43

 vantage

1.35

 perspectives

1.28

 viewpoint

1.19

 sidelines

1.09

 onset

1.03

 mouths

1.00

 beginnings

1.00

Activations Density 0.104%

phrases related to criticism or negative evaluations

phrases that emphasize the concept of beginnings or starting points

No Comments

No Known Activations