INDEX

Explanations

specific mentions of small objects or quantities

oai_token-act-pair · gpt-3.5-turbo

mentions of small or insignificant items and their accumulated significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Skies

-0.72

 Roses

-0.67

etheus

-0.66

andise

-0.65

hibited

-0.65

ukong

-0.63

Bridge

-0.63

 Rousse

-0.62

 Gleaming

-0.61

 Material

-0.60

POSITIVE LOGITS

(<

1.32

tiny

0.86

(~

0.78

 insignificant

0.78

pox

0.76

(>

0.76

 increments

0.76

 scattered

0.74

fry

0.74

iameter

0.72

Activations Density 0.354%

specific mentions of small objects or quantities

mentions of small or insignificant items and their accumulated significance

No Comments

No Known Activations