INDEX

Explanations

contractions specifically involving the words "don't" or "doesn't"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "don't."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 behavi

-0.81

ĪĴ

-0.76

Reviewer

-0.76

 Radiation

-0.70

 ventilation

-0.69

 antioxid

-0.68

ħĭ

-0.65

£ı

-0.65

 Penguin

-0.64

 Positive

-0.63

POSITIVE LOGITS

cha

1.15

ween

1.00

ople

0.96

otally

0.90

ting

0.89

unes

0.87

 necessarily

0.87

ional

0.84

acular

0.84

aper

0.83

Activations Density 0.091%

contractions specifically involving the words "don't" or "doesn't"

occurrences of the phrase "don't."

No Comments

No Known Activations