INDEX

Explanations

comparisons using the word "like"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "like"

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

ulty

-0.87

chin

-0.84

hiba

-0.84

inoa

-0.77

ourse

-0.77

Dispatch

-0.76

oard

-0.74

rax

-0.73

onte

-0.73

idates

-0.71

POSITIVE LOGITS

lihood

1.67

lier

1.02

liest

0.95

 ours

0.94

 minded

0.90

minded

0.90

liness

0.89

 wildfire

0.80

 hers

0.76

 yours

0.73

Activations Density 0.118%

comparisons using the word "like"

instances of the word "like"

No Comments

No Known Activations