INDEX

Explanations

personal opinions or reflections

oai_token-act-pair · gpt-3.5-turbo

personal pronouns and expressions of individual perspective

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

Loading

-0.74

Stock

-0.71

Availability

-0.70

Lot

-0.65

Appearances

-0.63

 kernels

-0.62

Detailed

-0.62

Spawn

-0.62

Output

-0.60

 dism

-0.60

POSITIVE LOGITS

 friends

0.73

 coworkers

0.72

 colleagues

0.69

å·

0.69

 myself

0.68

ACTED

0.66

usalem

0.65

friends

0.65

opic

0.65

 mates

0.65

Activations Density 0.425%

personal opinions or reflections

personal pronouns and expressions of individual perspective

No Comments

No Known Activations