INDEX

Explanations

references to wealth or financial success

oai_token-act-pair · gpt-3.5-turbo

references to wealth or financial success

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

oval

-0.76

ivity

-0.76

organic

-0.75

APD

-0.74

orney

-0.71

itarian

-0.71

itative

-0.68

INTON

-0.67

ells

-0.67

cible

-0.66

POSITIVE LOGITS

 fortune

1.11

 fortunes

0.95

 tremend

0.86

 parachute

0.73

 infl

0.71

 Seym

0.70

 stash

0.68

 earnings

0.68

awaru

0.67

 prosper

0.66

Activations Density 0.009%

references to wealth or financial success

references to wealth or financial success

No Comments

No Known Activations