INDEX

Explanations

dates mentioned in various formats and contexts

oai_token-act-pair · gpt-3.5-turbo

numerical dates and significant events associated with them

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

minecraft

-0.71

found

-0.68

cules

-0.66

ensical

-0.66

usercontent

-0.66

cientious

-0.64

cient

-0.60

abolic

-0.60

yrim

-0.60

netflix

-0.60

POSITIVE LOGITS

th

0.85

 deadline

0.70

 Rath

0.66

ãĤ´ãĥ³

0.65

 fault

0.65

 Memorial

0.64

ths

0.64

 onward

0.63

 onwards

0.63

enda

0.61

Activations Density 0.056%

dates mentioned in various formats and contexts

numerical dates and significant events associated with them

No Comments

No Known Activations