INDEX

Explanations

news articles discussing potential or likely scenarios, events, or actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "possible" indicating uncertainty or potentiality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

masters

-0.91

bane

-0.89

lua

-0.85

rix

-0.84

ogie

-0.83

ogun

-0.81

conom

-0.80

waters

-0.79

ien

-0.78

yer

-0.77

POSITIVE LOGITS

 defect

0.81

 embodiments

0.76

 feas

0.76

 inclusion

0.73

 implication

0.72

 demise

0.71

ROR

0.71

 alternatives

0.71

00000000

0.70

 surv

0.69

Activations Density 0.023%

news articles discussing potential or likely scenarios, events, or actions

instances of the word "possible" indicating uncertainty or potentiality

No Comments

No Known Activations