INDEX

Explanations

character sequences related to data analysis and computational tasks

oai_token-act-pair · gpt-3.5-turbo

topics related to environmental characteristics and measurements

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

nesty

-0.61

 humane

-0.60

 empowering

-0.57

å§

-0.54

 soothing

-0.54

 powerful

-0.53

 fraught

-0.53

 conven

-0.52

 unheard

-0.51

Nir

-0.51

POSITIVE LOGITS

 versus

0.82

vs

0.80

verages

0.80

 dataset

0.78

<-

0.73

Stats

0.68

 probabilities

0.68

->

0.68

->

0.67

 totals

0.65

Activations Density 0.770%

character sequences related to data analysis and computational tasks

topics related to environmental characteristics and measurements

No Comments

No Known Activations