INDEX

Explanations

mentions of the location "Brom" in various contexts

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to locations

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

 Medicaid

-0.70

WS

-0.67

IDENT

-0.65

ENCY

-0.65

USE

-0.64

 enrollment

-0.64

CEPT

-0.62

selection

-0.61

natureconservancy

-0.61

<+

-0.58

POSITIVE LOGITS

wich

1.52

 Brom

1.26

pton

1.02

mere

0.94

thal

0.89

bles

0.87

alties

0.86

sten

0.85

ania

0.85

elong

0.82

Activations Density 0.006%

mentions of the location "Brom" in various contexts

proper nouns related to locations

No Comments

No Known Activations