INDEX

Explanations

names, specifically the name "Marie"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names, particularly "Marie" and "Pierre"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ODUCT

-0.86

oted

-0.81

ORE

-0.79

otional

-0.76

ifiers

-0.76

itious

-0.75

rawling

-0.75

ifying

-0.74

atable

-0.73

arta

-0.73

POSITIVE LOGITS

 Claire

0.97

lla

0.86

 Slaughter

0.84

fen

0.78

lette

0.76

 Marie

0.72

 Louise

0.69

 Ples

0.67

bilt

0.66

Cur

0.66

Activations Density 0.017%

names, specifically the name "Marie"

mentions of specific names, particularly "Marie" and "Pierre"

No Comments

No Known Activations