INDEX

Explanations

phrases indicating a proposed action or suggestion

oai_token-act-pair · gpt-3.5-turbo

phrases that express obligation or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Fra

-0.66

atile

-0.64

 deteriorated

-0.63

CLR

-0.62

 Glob

-0.62

 adolesc

-0.60

ROS

-0.59

Kah

-0.57

 elusive

-0.57

 Puzzle

-0.56

POSITIVE LOGITS

 beware

1.01

be

0.97

nt

0.94

ered

0.93

 strive

0.92

erest

0.91

0.86

 aspire

0.84

've

0.83

 consider

0.83

Activations Density 0.068%

phrases indicating a proposed action or suggestion

phrases that express obligation or recommendations

No Comments

No Known Activations