INDEX

Explanations

phrases related to probable cause

oai_token-act-pair · gpt-3.5-turbo

phrases indicating likelihood or uncertainty

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

cock

-0.64

 Pool

-0.64

aven

-0.62

 Guides

-0.61

Layer

-0.61

 spotlight

-0.60

Anim

-0.58

roup

-0.58

ist

-0.57

ÂŃ

-0.57

POSITIVE LOGITS

 probable

3.99

bable

1.82

 plausible

1.44

 doubtful

1.33

 improbable

1.18

 conceivable

1.14

 practicable

1.14

 possible

1.11

 probability

1.08

 feasible

1.08

Activations Density 0.017%

phrases related to probable cause

phrases indicating likelihood or uncertainty

No Comments

No Known Activations