INDEX

Explanations

phrases indicating evidence or demonstration

oai_token-act-pair · gpt-3.5-turbo

phrases related to evidence and demonstration of claims or observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ieties

-0.74

ortium

-0.72

isers

-0.71

 warr

-0.67

shaw

-0.67

 dunno

-0.67

uthor

-0.66

 intending

-0.64

tic

-0.64

 prone

-0.63

POSITIVE LOGITS

 Figures

0.79

 Figure

0.76

 vividly

0.74

Fig

0.74

ADRA

0.71

 tangible

0.70

 Watts

0.68

products

0.68

Fig

0.66

 graphs

0.65

Activations Density 0.304%

phrases indicating evidence or demonstration

phrases related to evidence and demonstration of claims or observations

No Comments

No Known Activations