INDEX

Explanations

instances of people or things being seen or appearing in different contexts or locations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "in" followed by context or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

stroke

-0.69

ect

-0.67

ZA

-0.66

inion

-0.66

ente

-0.65

 PLEASE

-0.61

tips

-0.59

frog

-0.58

stein

-0.58

cum

-0.57

POSITIVE LOGITS

 action

1.07

 commercials

1.04

 photographs

1.02

 silhouette

0.99

 advertisements

0.98

 person

0.95

 movies

0.91

 theaters

0.89

action

0.89

ads

0.88

Activations Density 0.195%

instances of people or things being seen or appearing in different contexts or locations

occurrences of the phrase "in" followed by context or events

No Comments

No Known Activations