INDEX

Explanations

names, specifically the name "Jeffrey."

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

endars

-0.77

 Nuggets

-0.74

 Suns

-0.73

rum

-0.73

amus

-0.71

 Sync

-0.69

Jam

-0.67

 calendars

-0.67

sol

-0.66

isco

-0.64

POSITIVE LOGITS

 Jeffrey

2.40

uffer

1.96

 Scarborough

1.12

 Seymour

1.06

 Epstein

0.96

 Malk

0.96

vi

0.93

 Sidney

0.93

 Irwin

0.92

 Erik

0.89

Activations Density 0.055%

names, specifically the name "Jeffrey."

mentions of specific names and locations

No Comments

No Known Activations