INDEX

Explanations

instances of unexpected or surprising events or revelations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "twist" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

audi

-0.75

leased

-0.68

 redistributed

-0.66

 Hurricanes

-0.65

minent

-0.63

 Defenders

-0.63

hammad

-0.62

arel

-0.60

RES

-0.60

usable

-0.60

POSITIVE LOGITS

 twist

1.29

 twists

1.16

 Twist

0.91

 endings

0.88

 Whedon

0.83

 stitch

0.83

 weave

0.80

 twisting

0.79

stroke

0.77

nir

0.77

Activations Density 0.009%

instances of unexpected or surprising events or revelations

instances of the word "twist" in various contexts

No Comments

No Known Activations