INDEX

Explanations

mentions of locations or entities related to specific events or organizations

oai_token-act-pair · gpt-3.5-turbo

variations of the word "sense."

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

 Tata

-0.65

 Lazarus

-0.65

kees

-0.62

 Mahar

-0.62

RBI

-0.60

 known

-0.59

 cancelled

-0.59

BMC

-0.56

 resc

-0.56

 Abbott

-0.56

POSITIVE LOGITS

ense

4.83

enses

2.72

ENSE

2.49

ensing

1.98

ensed

1.82

ensive

1.55

enser

1.50

ensity

1.46

ENS

1.44

ence

1.44

Activations Density 0.011%

mentions of locations or entities related to specific events or organizations

variations of the word "sense."

No Comments

No Known Activations