INDEX

Explanations

dates written in a specific format - month and year separated by a comma

oai_token-act-pair · gpt-3.5-turbo

numerical sequences or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 exha

-0.70

 Ging

-0.65

 waivers

-0.63

NF

-0.62

 Graves

-0.61

 Fior

-0.60

 DRAG

-0.60

 assumptions

-0.60

 consulting

-0.59

 Kass

-0.59

POSITIVE LOGITS

0.97

0.95

0.93

0.92

chev

0.92

0.91

wm

0.90

0.89

Activations Density 0.102%

dates written in a specific format - month and year separated by a comma

numerical sequences or identifiers

No Comments

No Known Activations