INDEX

Explanations

mentions of financial penalties or fines

oai_token-act-pair · gpt-3.5-turbo

terms related to monetary penalties or fines

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

omorphic

-0.89

omorph

-0.89

groups

-0.80

ween

-0.77

ocaly

-0.74

rians

-0.74

ãĤ¤ãĥĪ

-0.72

OB

-0.71

space

-0.70

rina

-0.69

POSITIVE LOGITS

 levied

1.12

 imposed

0.85

 fines

0.83

 incurred

0.81

 fined

0.78

($

0.76

 hefty

0.75

 evasion

0.75

 penalty

0.75

 fine

0.74

Activations Density 0.022%

mentions of financial penalties or fines

terms related to monetary penalties or fines

No Comments

No Known Activations