INDEX

Explanations

mentions of events, laws, or initiatives being introduced

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "introduced" 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 @ 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

 Sioux

-0.67

die

-0.65

ergic

-0.64

istic

-0.63

 Templ

-0.62

opic

-0.61

 workspace

-0.60

 Myers

-0.60

 chance

-0.59

mic

-0.58

POSITIVE LOGITS

 introduced

1.16

introdu

1.06

 introduce

1.04

 introduces

0.97

 Introduced

0.91

 reintrodu

0.90

urated

0.89

 introducing

0.87

Introdu

0.84

ĸļ

0.83

Activations Density 0.010%

mentions of events, laws, or initiatives being introduced

occurrences of the word "introduced" in various contexts

No Comments

No Known Activations