INDEX

Explanations

dates mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Last" indicating time-related references in a sequence

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

bryce

-0.63

onto

-0.61

bsite

-0.59

adle

-0.57

cks

-0.57

okers

-0.56

=~

-0.55

plain

-0.53

ordes

-0.53

POSITIVE LOGITS

 week

1.17

 month

1.14

 Updated

1.14

 year

1.05

 night

0.99

 updated

0.95

 Update

0.91

 summer

0.90

 Month

0.90

 weekend

0.89

Activations Density 0.037%

dates mentioned in a text

occurrences of the word "Last" indicating time-related references in a sequence

No Comments

No Known Activations