INDEX

Explanations

mentions of dates and locations, potentially related to events, publications, or people

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and their frequency in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 desperation

-0.83

 maybe

-0.81

 favour

-0.79

 temptation

-0.78

 imagination

-0.78

 reflex

-0.77

cra

-0.77

 bending

-0.76

 civilisation

-0.76

 sometimes

-0.75

POSITIVE LOGITS

Tickets

1.59

Among

1.57

According

1.56

Additionally

1.53

Prior

1.49

Laun

1.48

Each

1.48

RELATED

1.48

The

1.47

Along

1.44

Activations Density 0.489%

mentions of dates and locations, potentially related to events, publications, or people

punctuation marks and their frequency in sentences

No Comments

No Known Activations