INDEX

Explanations

names of notable individuals along with their respective ages and professional affiliations

oai_token-act-pair · gpt-3.5-turbo

punctuated endings of character names or references

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

aban

-0.71

hor

-0.67

kus

-0.65

chin

-0.65

iku

-0.65

ton

-0.65

TON

-0.63

rogen

-0.63

rosis

-0.63

arna

-0.61

POSITIVE LOGITS

 namely

0.69

 Flavoring

0.68

————

0.67

channelAvailability

0.64

 meanwhile

0.63

 however

0.63

————————

0.61

 Neph

0.60

 nevertheless

0.60

anship

0.59

Activations Density 0.086%

names of notable individuals along with their respective ages and professional affiliations

punctuated endings of character names or references

No Comments

No Known Activations