INDEX

Explanations

comma-separated elements in a list

oai_token-act-pair · gpt-3.5-turbo

instances of personal pronouns and punctuation indicating speech or dialogue

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

 elusive

-0.72

 viability

-0.68

 attractiveness

-0.65

 adoption

-0.64

 Landing

-0.63

lif

-0.63

 kidnapping

-0.62

 rendition

-0.62

ORN

-0.62

 abduction

-0.61

POSITIVE LOGITS

essed

0.84

aca

0.74

felt

0.72

lived

0.70

pires

0.70

stood

0.69

ided

0.69

might

0.69

uld

0.69

should

0.68

Activations Density 0.206%

comma-separated elements in a list

instances of personal pronouns and punctuation indicating speech or dialogue

No Comments

No Known Activations