INDEX

Explanations

mentions of different types of boots

oai_token-act-pair · gpt-3.5-turbo

references to various types of boots

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Topics

-0.81

SHIP

-0.74

MRI

-0.71

ulton

-0.69

 Plasma

-0.69

 Neurolog

-0.68

 Neuro

-0.66

 Hert

-0.66

Ĭ±

-0.66

 Huntington

-0.64

POSITIVE LOGITS

 boots

1.51

 Boots

1.10

strap

1.05

 shoes

0.95

 gloves

0.93

 footwear

0.91

 sleeves

0.87

 heel

0.87

 socks

0.86

 boot

0.84

Activations Density 0.004%

mentions of different types of boots

references to various types of boots

No Comments

No Known Activations