INDEX

Explanations

mentions of the city "Des Moines."

oai_token-act-pair · gpt-3.5-turbo

references to geographic locations and associated entities

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

fired

-0.65

 sequest

-0.65

 Clicker

-0.62

 restraints

-0.61

holding

-0.61

--------------------------------------------------------

-0.61

saf

-0.61

binary

-0.61

 compuls

-0.60

 Finder

-0.60

POSITIVE LOGITS

 Moines

1.03

erville

0.86

ippi

0.86

hyde

0.86

 Register

0.85

oxide

0.75

anism

0.74

ktop

0.74

 Timberwolves

0.73

stad

0.72

Activations Density 0.051%

mentions of the city "Des Moines."

references to geographic locations and associated entities

No Comments

No Known Activations