INDEX

Explanations

phrases related to ownership or affiliation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating organizational affiliations or hierarchical structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

fw

-0.80

 warranted

-0.77

epad

-0.75

apy

-0.69

 dictated

-0.68

hyde

-0.68

matter

-0.66

reci

-0.65

 outweigh

-0.63

 await

-0.63

POSITIVE LOGITS

 sorts

0.80

 Britain

0.73

 famed

0.73

lishes

0.70

rium

0.70

 disgr

0.68

?)

0.68

 Belarus

0.66

 former

0.64

 Lloyd

0.64

Activations Density 0.316%

phrases related to ownership or affiliation

phrases indicating organizational affiliations or hierarchical structures

No Comments

No Known Activations