INDEX

Explanations

words related to cleanliness and organization

oai_token-act-pair · gpt-3.5-turbo

variations of the prefix "cl" and its associated words

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

 tremend

-0.71

 unexpl

-0.70

 latent

-0.70

 discriminate

-0.69

 discrimination

-0.68

 warr

-0.67

Magikarp

-0.66

 goodwill

-0.65

OTOS

-0.64

ONSORED

-0.63

POSITIVE LOGITS

nces

0.86

heed

0.86

ico

0.79

hemy

0.78

nant

0.78

mont

0.77

*/(

0.77

onica

0.76

veland

0.75

ulhu

0.73

Activations Density 0.077%

words related to cleanliness and organization

variations of the prefix "cl" and its associated words

No Comments

No Known Activations