INDEX

Explanations

phrases related to falling or being off track/down

oai_token-act-pair · gpt-3.5-turbo

instances of personal transformation or emotional experiences

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

 prominently

-0.81

 roundup

-0.79

 Offic

-0.71

 cheaply

-0.69

SourceFile

-0.67

 reportedly

-0.64

expensive

-0.64

 proudly

-0.63

 inspected

-0.63

Thousands

-0.61

POSITIVE LOGITS

 conclusions

1.01

illusion

0.97

enment

0.92

EStreamFrame

0.91

thinking

0.89

izons

0.83

athetic

0.78

 toes

0.78

 understanding

0.77

 direction

0.77

Activations Density 0.406%

phrases related to falling or being off track/down

instances of personal transformation or emotional experiences

No Comments

No Known Activations