INDEX

Explanations

references to a fork as a physical object

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "fork" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

useum

-0.73

Yao

-0.67

 Danger

-0.66

MPG

-0.66

 Madden

-0.63

 Origins

-0.63

 Neville

-0.62

zona

-0.61

istics

-0.61

 unaccompanied

-0.60

POSITIVE LOGITS

lift

1.54

edIn

0.93

crew

0.91

nown

0.86

ingen

0.85

ners

0.84

walk

0.83

hire

0.82

pit

0.79

istani

0.79

Activations Density 0.058%

references to a fork as a physical object

occurrences of the word "fork" in various contexts

No Comments

No Known Activations