INDEX

Explanations

keywords related to pinpointing, spotting, recognizing, and identifying specific information or features

oai_token-act-pair · gpt-3.5-turbo

terms related to detection and identification processes

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

udeb

-0.73

cloth

-0.71

Sov

-0.71

adra

-0.69

ylum

-0.69

undy

-0.67

usp

-0.66

rette

-0.65

orld

-0.65

HOU

-0.63

POSITIVE LOGITS

alys

0.86

ibly

0.82

 detect

0.81

 Detect

0.80

 detecting

0.79

pell

0.76

 neural

0.75

 accurately

0.75

 accuracy

0.74

 Accuracy

0.73

Activations Density 0.288%

keywords related to pinpointing, spotting, recognizing, and identifying specific information or features

terms related to detection and identification processes

No Comments

No Known Activations