INDEX

Explanations

phrases related to facing challenges or negative outcomes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant actions or changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ndra

-0.76

orth

-0.75

bably

-0.71

osity

-0.70

-+-+

-0.70

ulty

-0.70

arth

-0.67

anni

-0.67

legates

-0.66

apo

-0.66

POSITIVE LOGITS

 seriously

0.98

 lightly

0.87

 aback

0.80

 virginity

0.80

 hostage

0.78

 Seriously

0.77

 stride

0.77

 plunge

0.76

cue

0.76

 reins

0.75

Activations Density 0.255%

phrases related to facing challenges or negative outcomes

phrases indicating significant actions or changes

No Comments

No Known Activations