INDEX

Explanations

instances where a measurement concept "span" is mentioned in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "span" and its variations indicating range or duration

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

ysis

-0.78

ertodd

-0.74

phis

-0.70

tarians

-0.68

oyal

-0.67

ainment

-0.67

tical

-0.66

tics

-0.66

illard

-0.65

 aspir

-0.62

POSITIVE LOGITS

ning

1.10

ming

0.89

span

0.88

ner

0.88

ners

0.84

 span

0.82

ider

0.81

 Span

0.81

 spanning

0.79

vier

0.76

Activations Density 0.017%

instances where a measurement concept "span" is mentioned in various contexts

occurrences of the word "span" and its variations indicating range or duration

No Comments

No Known Activations