INDEX

Explanations

proper nouns, specifically names, particularly focusing on the name "Leonard"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Leonard."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yrim

-0.81

skirts

-0.77

illance

-0.76

manship

-0.69

tch

-0.68

Ã±

-0.68

pron

-0.67

dule

-0.65

eden

-0.65

usp

-0.63

POSITIVE LOGITS

 Leonard

1.45

 Lauder

0.93

 Cohen

0.81

angelo

0.78

Angelo

0.77

 Strauss

0.75

 Bridges

0.73

ergic

0.73

Nim

0.72

ius

0.72

Activations Density 0.006%

proper nouns, specifically names, particularly focusing on the name "Leonard"

mentions of the name "Leonard."

No Comments

No Known Activations