INDEX

Explanations

phrases indicating clarity or decisiveness

oai_token-act-pair · gpt-3.5-turbo

phrases indicating clarity and straightforwardness

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tremend

-0.76

 Loft

-0.68

 nostalg

-0.66

uld

-0.65

 awkwardly

-0.64

cture

-0.61

inse

-0.60

cule

-0.60

FactoryReloaded

-0.59

 therap

-0.59

POSITIVE LOGITS

ances

1.41

cut

1.34

headed

1.14

ance

1.10

cutting

1.04

cuts

0.99

 deline

0.96

 indication

0.92

 implication

0.84

iary

0.80

Activations Density 0.050%

phrases indicating clarity or decisiveness

phrases indicating clarity and straightforwardness

No Comments

No Known Activations