INDEX

Explanations

the word "Spot" in the text

oai_token-act-pair · gpt-3.5-turbo

references to "Spot" and its variations in context, particularly in relation to movies or media

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

KS

-0.75

externalActionCode

-0.74

 derog

-0.72

 mortg

-0.68

KI

-0.67

Erd

-0.67

 oath

-0.66

 descendant

-0.66

ste

-0.65

 chars

-0.63

POSITIVE LOGITS

Spot

3.86

spot

3.28

 Spot

2.82

 Spotify

1.67

 spot

1.63

 Spotlight

1.60

 spots

1.40

 spotting

1.35

Sensor

0.94

 Tracks

0.90

Activations Density 0.012%

the word "Spot" in the text

references to "Spot" and its variations in context, particularly in relation to movies or media

No Comments

No Known Activations