INDEX

Explanations

phrases starting with "Two" followed by a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Two"

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

Ö

-0.83

 guiActiveUnfocused

-0.74

Ïī

-0.65

zyme

-0.65

ifice

-0.64

Ú

-0.62

 widest

-0.62

||||

-0.61

ãĤ¦

-0.61

ocratic

-0.61

POSITIVE LOGITS

teen

1.09

een

1.04

 Ways

0.97

 Months

0.91

 Weeks

0.86

resa

0.83

eteen

0.80

Thirty

0.79

 Strikes

0.79

 Hundred

0.79

Activations Density 0.054%

phrases starting with "Two" followed by a number

occurrences of the word "Two"

No Comments

No Known Activations