INDEX

Explanations

terms related to awards or recognition, particularly the phrase 'Cross'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Cross."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

UCH

-0.74

acious

-0.74

ailable

-0.73

 "$:/

-0.73

 WATCHED

-0.73

ENCY

-0.72

OST

-0.70

EMBER

-0.68

*/(

-0.67

ETHOD

-0.66

POSITIVE LOGITS

roads

1.27

bones

1.15

Cross

0.96

bows

0.95

fax

0.94

bow

0.92

cross

0.88

 Cross

0.87

hurst

0.84

es

0.84

Activations Density 0.019%

terms related to awards or recognition, particularly the phrase 'Cross'

occurrences of the word "Cross."

No Comments

No Known Activations