INDEX

Explanations

mentions of specific names, possibly related to politics and events

oai_token-act-pair · gpt-3.5-turbo

punctuation and names, particularly those related to people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

UFF

-0.87

Weiss

-0.84

ocks

-0.81

DDR

-0.79

 dere

-0.76

verett

-0.74

 wolves

-0.73

~~~~~~~~

-0.73

 Norn

-0.73

uff

-0.72

POSITIVE LOGITS

Pal

1.97

Pal

1.83

pal

1.61

 Palmer

1.51

pal

1.47

PAL

1.45

Cal

1.21

Kal

1.20

Kal

1.19

 Palm

1.17

Activations Density 0.288%

mentions of specific names, possibly related to politics and events

punctuation and names, particularly those related to people

No Comments

No Known Activations