INDEX

Explanations

phrases or keywords related to eligibility for different programs or benefits

oai_token-act-pair · gpt-3.5-turbo

phrases indicating eligibility for various programs or benefits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ijk

-0.68

Else

-0.67

rough

-0.65

tons

-0.60

chief

-0.60

stead

-0.60

ERN

-0.59

 Sabb

-0.58

il

-0.58

aida

-0.58

POSITIVE LOGITS

untarily

0.96

igible

0.84

WP

0.75

iaries

0.74

iciary

0.72

iary

0.71

ioned

0.71

 qualify

0.70

 eligible

0.69

mma

0.69

Activations Density 0.041%

phrases or keywords related to eligibility for different programs or benefits

phrases indicating eligibility for various programs or benefits

No Comments

No Known Activations