INDEX

Explanations

phrases involving the number "one" with specific context or other related elements

oai_token-act-pair · gpt-3.5-turbo

references to the number two in various contexts

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

ocent

-0.75

acements

-0.74

ury

-0.70

arcity

-0.69

stru

-0.69

 Shutterstock

-0.68

models

-0.66

lov

-0.64

politics

-0.64

renches

-0.64

POSITIVE LOGITS

 apiece

0.79

 halves

0.77

 accuser

0.67

 side

0.67

 iteration

0.64

ngth

0.63

ERA

0.62

 sided

0.60

 camel

0.59

 Learns

0.57

Activations Density 0.098%

phrases involving the number "one" with specific context or other related elements

references to the number two in various contexts

No Comments

No Known Activations