INDEX

Explanations

phrases indicating visual observations or sightings

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seen."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

angan

-0.71

vernment

-0.70

 unification

-0.64

cussion

-0.64

orrect

-0.63

 anthem

-0.63

 misunderstanding

-0.61

 Mercenary

-0.60

EStreamFrame

-0.59

 circumstance

-0.59

POSITIVE LOGITS

ById

0.93

IRT

0.89

IOR

0.87

Ĺ

0.86

dust

0.81

wcsstore

0.81

PsyNetMessage

0.81

ĸ

0.78

Model

0.76

seeing

0.75

Activations Density 0.028%

phrases indicating visual observations or sightings

instances of the word "seen."

No Comments

No Known Activations