INDEX

Explanations

words related to enabling or disabling settings or features

oai_token-act-pair · gpt-3.5-turbo

phrases related to enablement and configuration settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

apest

-0.95

ivan

-0.75

letes

-0.75

orians

-0.75

sheets

-0.75

ourcing

-0.75

Catal

-0.74

von

-0.74

ographers

-0.74

forts

-0.73

POSITIVE LOGITS

 threshold

0.84

 override

0.81

 enabled

0.81

 automatic

0.80

 boolean

0.79

 mode

0.78

 CONFIG

0.78

 integer

0.77

 defaults

0.76

 parity

0.75

Activations Density 0.234%

words related to enabling or disabling settings or features

phrases related to enablement and configuration settings

No Comments

No Known Activations