INDEX

Explanations

references to a specific entity or concept within a larger context

oai_token-act-pair · gpt-3.5-turbo

repeated phrases and word patterns emphasizing "all" and "the."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ÃĥÃĤ

-0.79

elist

-0.74

odcast

-0.73

eller

-0.72

chard

-0.70

ister

-0.68

ello

-0.67

SPONSORED

-0.67

ename

-0.67

opsis

-0.65

POSITIVE LOGITS

 facets

0.89

 fuss

0.89

 goodies

0.87

 pent

0.84

 hoop

0.81

 bells

0.78

 complexities

0.76

 possibilities

0.74

 hoops

0.74

 glory

0.73

Activations Density 0.092%

references to a specific entity or concept within a larger context

repeated phrases and word patterns emphasizing "all" and "the."

No Comments

No Known Activations