INDEX

Explanations

names of specific hotels and venues

oai_token-act-pair · gpt-3.5-turbo

references to specific names and brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

qua

-0.98

å£«

-0.94

zynski

-0.77

cles

-0.77

gary

-0.75

leigh

-0.75

pport

-0.74

FTWARE

-0.74

Ry

-0.71

xual

-0.70

POSITIVE LOGITS

uate

0.90

isf

0.76

uated

0.74

abal

0.69

pad

0.67

ippi

0.67

ongh

0.66

uating

0.66

 tutor

0.63

uates

0.63

Activations Density 0.098%

names of specific hotels and venues

references to specific names and brands

No Comments

No Known Activations