INDEX

Explanations

words related to containers, specifically jars of various kinds

oai_token-act-pair · gpt-3.5-turbo

references to jars

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

issance

-0.99

âĸ¬

-0.78

 antioxid

-0.78

deen

-0.76

Club

-0.71

 dismant

-0.68

Edge

-0.67

vous

-0.67

 Explicit

-0.67

 mosqu

-0.66

POSITIVE LOGITS

iances

0.92

atory

0.89

pole

0.77

atories

0.74

chuk

0.70

atown

0.70

anyahu

0.70

PTS

0.66

oulos

0.66

endum

0.65

Activations Density 0.040%

words related to containers, specifically jars of various kinds

references to jars

No Comments

No Known Activations