INDEX

Explanations

references to chocolate products

oai_token-act-pair · gpt-3.5-turbo

mentions of chocolate

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

atives

-0.81

WARD

-0.78

umbnail

-0.76

abilities

-0.70

seless

-0.70

 plur

-0.69

href

-0.68

ivot

-0.66

 REPORT

-0.65

 Fargo

-0.65

POSITIVE LOGITS

 chocolate

0.97

 cake

0.97

 pudding

0.96

 coated

0.88

 flavored

0.87

 syrup

0.86

anut

0.85

 flav

0.85

 cane

0.82

âĺħâĺħ

0.82

Activations Density 0.014%

references to chocolate products

mentions of chocolate

No Comments

No Known Activations