INDEX

Explanations

phrases related to locations and statistics

oai_token-act-pair · gpt-3.5-turbo

concepts related to statistics and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

terness

-0.87

rocal

-0.84

vind

-0.82

Å¡

-0.71

arre

-0.71

zin

-0.70

rils

-0.67

nces

-0.67

ouble

-0.67

resy

-0.67

POSITIVE LOGITS

 EVER

0.79

 ever

0.78

 imaginable

0.68

amen

0.68

 humankind

0.66

 mankind

0.66

aceae

0.64

earch

0.63

 modern

0.62

 contemporary

0.62

Activations Density 0.380%

phrases related to locations and statistics

concepts related to statistics and their implications

No Comments

No Known Activations