INDEX

Explanations

references to a particular term, "Shin."

oai_token-act-pair · gpt-3.5-turbo

references to the term "Shin," likely related to a specific organization or concept

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

 Hurricanes

-0.78

Pwr

-0.74

ENC

-0.74

address

-0.73

nces

-0.71

utics

-0.71

ENCE

-0.71

 Mayweather

-0.69

ĳ

-0.66

aneers

-0.66

POSITIVE LOGITS

Shin

1.20

 Shin

1.12

obi

0.95

omi

0.87

ook

0.80

 Nanto

0.80

 seiz

0.79

 shin

0.79

ning

0.76

etsu

0.75

Activations Density 0.005%

references to a particular term, "Shin."

references to the term "Shin," likely related to a specific organization or concept

No Comments

No Known Activations

references to a particular term, "Shin."

references to the term "Shin," likely related to a specific organization or concept

No Comments

No Known Activations