INDEX

Explanations

adjectives describing extremes or intensities

oai_token-act-pair · gpt-3.5-turbo

adjectives indicating extremes or notable characteristics

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

mun

-0.82

skirts

-0.74

è£ħ

-0.72

Þ

-0.71

 teasp

-0.70

clair

-0.69

bara

-0.69

Lyn

-0.67

isha

-0.67

ADS

-0.66

POSITIVE LOGITS

 imaginable

1.11

 thing

0.92

 manifestation

0.91

 predictor

0.81

 conceivable

0.80

 exponent

0.76

 culprit

0.76

 ever

0.75

hest

0.73

 possible

0.73

Activations Density 0.097%

adjectives describing extremes or intensities

adjectives indicating extremes or notable characteristics

No Comments

No Known Activations