INDEX

Explanations

personal pronouns followed by the action of telling information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating reported speech or quotations

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

nets

-0.67

earable

-0.67

 tides

-0.67

otine

-0.66

Catalog

-0.64

profits

-0.62

tin

-0.60

herence

-0.60

orsche

-0.60

JUST

-0.59

POSITIVE LOGITS

 eleph

0.76

racuse

0.76

cca

0.76

adow

0.73

ldon

0.72

 downstairs

0.71

zzo

0.69

adows

0.68

çīĪ

0.67

imei

0.66

Activations Density 0.074%

personal pronouns followed by the action of telling information

phrases indicating reported speech or quotations

No Comments

No Known Activations