INDEX

Explanations

past tense verbs ending in 'ed'

oai_token-act-pair · gpt-3.5-turbo

terms related to classification and categorization

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

perty

-0.77

irgin

-0.76

sburg

-0.72

burg

-0.71

enegger

-0.66

issance

-0.64

udeb

-0.63

onda

-0.63

 Guth

-0.62

ighth

-0.62

POSITIVE LOGITS

naire

0.78

enance

0.74

 probabilities

0.72

 criteria

0.70

istics

0.69

opers

0.69

 Procedures

0.67

otomy

0.66

 geographically

0.66

 algorithms

0.66

Activations Density 0.158%

past tense verbs ending in 'ed'

terms related to classification and categorization

No Comments

No Known Activations