INDEX

Explanations

phrases related to numerical values and categories related to quantity, plurality, or categorization

oai_token-act-pair · gpt-3.5-turbo

terms related to political or structural frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.70

urga

-0.68

latest

-0.65

 anew

-0.64

ibaba

-0.64

heit

-0.64

ename

-0.63

ETA

-0.63

wolves

-0.63

\<

-0.60

POSITIVE LOGITS

 configurations

0.97

imensional

0.95

 setups

0.83

 configuration

0.77

SLI

0.74

edly

0.73

 affair

0.73

 arrangement

0.70

 occupancy

0.70

âĦ¢:

0.69

Activations Density 0.159%

phrases related to numerical values and categories related to quantity, plurality, or categorization

terms related to political or structural frameworks

No Comments

No Known Activations