INDEX

Explanations

terms related to accessibility

oai_token-act-pair · gpt-3.5-turbo

references to accessibility in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

hair

-0.66

roo

-0.65

ugh

-0.64

oil

-0.63

coat

-0.63

aleb

-0.63

ril

-0.62

uh

-0.61

etti

-0.61

nings

-0.60

POSITIVE LOGITS

 accessible

3.52

accessible

2.79

 inaccessible

2.33

 accessibility

2.04

 readable

1.60

 accessed

1.60

 usable

1.50

 available

1.42

 achievable

1.39

 habitable

1.32

Activations Density 0.015%

terms related to accessibility

references to accessibility in various contexts

No Comments

No Known Activations