INDEX

Explanations

phrases related to flags

oai_token-act-pair · gpt-3.5-turbo

references to various types of flags

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 mutual

-0.66

 medicine

-0.61

URRENT

-0.61

 simultaneous

-0.58

 LIFE

-0.58

ccording

-0.58

 reconstruction

-0.58

uclear

-0.57

 embodiment

-0.57

izen

-0.57

POSITIVE LOGITS

ags

1.04

gers

0.91

aws

0.85

ging

0.84

hi

0.84

gery

0.83

zag

0.81

glers

0.81

ged

0.80

heet

0.79

Activations Density 0.003%

phrases related to flags

references to various types of flags

No Comments

No Known Activations