INDEX

Explanations

proper nouns or names, especially related to people

oai_token-act-pair · gpt-3.5-turbo

names and terms related to specific people and places

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

TAG

-0.89

ĨĴ

-0.83

URES

-0.72

NECT

-0.72

natureconservancy

-0.72

ccording

-0.72

ĻĤ

-0.72

é¾

-0.67

Preview

-0.66

 faculties

-0.65

POSITIVE LOGITS

ktop

0.78

abase

0.77

estone

0.72

shire

0.67

ale

0.66

hy

0.66

illet

0.66

 Tale

0.63

ise

0.62

hus

0.61

Activations Density 0.312%

proper nouns or names, especially related to people

names and terms related to specific people and places

No Comments

No Known Activations