INDEX

Explanations

names with the prefix "f"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'f'

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

 paraly

-0.73

 Boone

-0.70

 machine

-0.67

 Lizard

-0.66

 theater

-0.65

 clipboard

-0.64

 chamber

-0.61

eer

-0.61

 Clash

-0.61

 colleg

-0.61

POSITIVE LOGITS

ertility

1.27

ruits

1.26

avorable

1.22

rozen

1.21

actory

1.19

ornia

1.17

eatured

1.16

rost

1.16

urther

1.15

raction

1.12

Activations Density 0.034%

names with the prefix "f"

occurrences of the letter 'f'

No Comments

No Known Activations