INDEX

Explanations

mentions of mothers

oai_token-act-pair · gpt-3.5-turbo

references to "mother."

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

 Flavoring

-0.93

NRS

-0.78

okers

-0.74

idian

-0.72

EY

-0.71

raviolet

-0.67

vernment

-0.67

urat

-0.67

ratulations

-0.66

ickr

-0.66

POSITIVE LOGITS

hood

1.12

hesis

1.09

heses

1.04

hetical

0.97

ship

0.97

hetically

0.95

load

0.91

ships

0.83

maid

0.82

 Teresa

0.81

Activations Density 0.040%

mentions of mothers

references to "mother."

No Comments

No Known Activations