INDEX

Explanations

dates mentioned in different contexts or events

oai_token-act-pair · gpt-3.5-turbo

references to specific years and events

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

amy

-0.85

abwe

-0.70

adata

-0.69

uden

-0.68

terday

-0.68

cloth

-0.64

phy

-0.64

ansk

-0.64

vier

-0.64

thia

-0.63

POSITIVE LOGITS

 edition

1.06

å¹

0.90

 census

0.84

 iteration

0.84

 version

0.84

0.83

 installment

0.83

 incarnation

0.81

 election

0.80

 inaugural

0.76

Activations Density 0.068%

dates mentioned in different contexts or events

references to specific years and events

No Comments

No Known Activations