INDEX

Explanations

phrases related to statistics or program outcomes

oai_token-act-pair · gpt-3.5-turbo

periods at the end of sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

IRE

-0.74

ŃĶ

-0.70

 fucked

-0.64

 haul

-0.63

ãĥĺ

-0.63

 libel

-0.63

 reven

-0.61

 racket

-0.61

zo

-0.59

tyr

-0.59

POSITIVE LOGITS

 Additionally

1.63

 Furthermore

1.54

 Conversely

1.48

 Moreover

1.48

 However

1.47

 Consequently

1.44

 Therefore

1.41

 Likewise

1.40

 Similarly

1.39

 Nevertheless

1.31

Activations Density 0.598%

phrases related to statistics or program outcomes

periods at the end of sentences

No Comments

No Known Activations