INDEX

Explanations

comparative adjectives ending in '-er'

oai_token-act-pair · gpt-3.5-turbo

comparisons and statements expressing improvement or superiority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Py

-0.74

 Pione

-0.68

NH

-0.66

idon

-0.66

FK

-0.64

erity

-0.64

aly

-0.62

mad

-0.62

TRY

-0.62

EP

-0.62

POSITIVE LOGITS

 than

1.08

 suited

1.07

 behaved

0.95

 acquainted

0.83

 Than

0.76

 quality

0.76

 manag

0.75

than

0.75

 predictor

0.71

 vers

0.70

Activations Density 0.033%

comparative adjectives ending in '-er'

comparisons and statements expressing improvement or superiority

No Comments

No Known Activations