INDEX

Explanations

the name "Rodgers" mentioned in sentences

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual, likely a sports figure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

uate

-0.88

uates

-0.73

izations

-0.72

uated

-0.72

 horizon

-0.66

uating

-0.64

 Liberties

-0.62

 downside

-0.62

ting

-0.62

 coordination

-0.61

POSITIVE LOGITS

 Rodgers

1.04

mith

0.80

manship

0.80

sey

0.79

 Yards

0.77

etta

0.76

adier

0.76

cliffe

0.75

gers

0.75

bey

0.74

Activations Density 0.005%

the name "Rodgers" mentioned in sentences

references to a specific individual, likely a sports figure

No Comments

No Known Activations