INDEX

Explanations

the term "post-human"

oai_token-act-pair · gpt-3.5-turbo

references to "human" or related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rolls

-0.81

 flats

-0.72

 avail

-0.71

 rolled

-0.67

 roll

-0.66

 premiums

-0.66

tie

-0.65

 subscriptions

-0.65

reb

-0.65

 filed

-0.64

POSITIVE LOGITS

human

4.02

humans

2.88

Human

2.45

 human

1.82

 Human

1.75

humane

1.67

 humans

1.52

 Humans

1.45

anthrop

1.39

Hum

1.32

Activations Density 0.014%

the term "post-human"

references to "human" or related concepts

No Comments

No Known Activations