INDEX

Explanations

references to well-being and related concepts

oai_token-act-pair · gpt-3.5-turbo

references to well-being

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NRS

-0.69

 Oops

-0.67

oslov

-0.66

PsyNetMessage

-0.65

 Wick

-0.65

sonian

-0.60

 peak

-0.59

ulla

-0.59

 fold

-0.58

 Lilly

-0.58

POSITIVE LOGITS

 able

0.89

quished

0.89

 Able

0.75

ibe

0.73

hemoth

0.72

terness

0.72

omorph

0.72

 eaten

0.71

peria

0.71

Angelo

0.71

Activations Density 0.010%

references to well-being and related concepts

references to well-being

No Comments

No Known Activations