INDEX

Explanations

large numerical values associated with quantities or statistics

oai_token-act-pair · gpt-3.5-turbo

repeated numeric values representing large quantities

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

 matt

-0.68

XD

-0.62

jo

-0.59

HC

-0.58

 marqu

-0.55

 plat

-0.55

nep

-0.53

:(

-0.53

 Finder

-0.52

 behav

-0.51

POSITIVE LOGITS

1.50

1.25

1.23

1.14

1.12

1.10

1.09

1.02

0.98

Activations Density 0.076%

large numerical values associated with quantities or statistics

repeated numeric values representing large quantities

No Comments

No Known Activations