INDEX

Explanations

twists or unexpected developments in a story or situation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "twist" along with its variations, often in contexts indicating unexpected changes or turns in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

leased

-0.70

audi

-0.69

hammad

-0.68

RES

-0.67

 Lama

-0.66

 Defenders

-0.65

eworld

-0.64

 Continued

-0.64

avement

-0.64

rahim

-0.63

POSITIVE LOGITS

 twist

1.20

 twists

1.17

 Twist

0.88

 endings

0.86

 Whedon

0.80

stroke

0.79

 twisting

0.77

abouts

0.72

 stitch

0.71

creen

0.70

Activations Density 0.012%

twists or unexpected developments in a story or situation

instances of the word "twist" along with its variations, often in contexts indicating unexpected changes or turns in narratives

No Comments

No Known Activations