INDEX

Explanations

references to a specific company, Verizon Wireless

oai_token-act-pair · gpt-3.5-turbo

mentions of the Verizon brand and its wireless services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

etic

-0.84

rophe

-0.83

etically

-0.78

etics

-0.77

obar

-0.77

anchester

-0.75

ovych

-0.75

vre

-0.75

eworld

-0.75

oly

-0.74

POSITIVE LOGITS

 Wireless

1.13

Fi

0.96

 Communications

0.91

izon

0.82

 Bros

0.78

ITE

0.75

 Laboratories

0.72

ISION

0.70

GE

0.69

 Corporation

0.68

Activations Density 0.016%

references to a specific company, Verizon Wireless

mentions of the Verizon brand and its wireless services

No Comments

No Known Activations