INDEX

Explanations

first person statements expressing personal opinions, actions, or experiences

oai_token-act-pair · gpt-3.5-turbo

reflections on personal identity and social responsibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 respectively

-0.74

 themselves

-0.73

 apiece

-0.61

 omin

-0.58

EMS

-0.57

 alike

-0.56

gerald

-0.54

Their

-0.53

 tariffs

-0.53

 Toll

-0.53

POSITIVE LOGITS

 myself

1.35

my

0.91

poke

0.79

oan

0.64

ograp

0.63

opic

0.62

 writing

0.62

 blogging

0.61

 personally

0.61

chin

0.59

Activations Density 0.961%

first person statements expressing personal opinions, actions, or experiences

reflections on personal identity and social responsibility

No Comments

No Known Activations