INDEX

Explanations

Roman numerals indicating the number 2 at the end of phrases

oai_token-act-pair · gpt-3.5-turbo

references to the Roman numeral "II" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 waking

-0.66

acity

-0.63

Pub

-0.63

uckland

-0.62

 pupp

-0.60

astical

-0.59

broad

-0.58

scribed

-0.58

ãĤ¦

-0.58

elong

-0.58

POSITIVE LOGITS

II

3.52

III

2.78

II

2.20

IV

2.09

VII

2.00

 VIII

1.90

III

1.87

XII

1.77

IX

1.71

VI

1.66

Activations Density 0.012%

Roman numerals indicating the number 2 at the end of phrases

references to the Roman numeral "II" in various contexts

No Comments

No Known Activations