INDEX

Explanations

words related to accessibility or ways things can be accessed

oai_token-act-pair · gpt-3.5-turbo

concepts related to accessibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

stal

-0.75

origin

-0.69

ortium

-0.66

chief

-0.64

neg

-0.63

²¾

-0.63

 squad

-0.62

aldi

-0.62

inate

-0.61

hon

-0.61

POSITIVE LOGITS

ibility

1.10

ible

0.99

ibly

0.83

accessible

0.80

 portals

0.74

 accessible

0.74

idges

0.72

[*

0.70

DOS

0.69

atable

0.67

Activations Density 0.024%

words related to accessibility or ways things can be accessed

concepts related to accessibility

No Comments

No Known Activations