INDEX

Explanations

mentions of marine or sea-related terms

oai_token-act-pair · gpt-3.5-turbo

references to the sea or ocean

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 denomin

-0.71

 favor

-0.70

CU

-0.65

 ellipt

-0.64

 prof

-0.63

Fet

-0.60

 transgress

-0.60

 programming

-0.59

 fest

-0.58

 mark

-0.58

POSITIVE LOGITS

sea

4.80

Sea

2.06

SEA

1.53

sea

1.50

Sea

1.41

marine

1.27

 Seas

1.20

 seaf

1.19

Ocean

1.19

worm

1.18

Activations Density 0.009%

mentions of marine or sea-related terms

references to the sea or ocean

No Comments

No Known Activations