INDEX

Explanations

dates or time-related phrases

oai_token-act-pair · gpt-3.5-turbo

references to a specific temporal context or timeframe

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

-0.72

chester

-0.72

atche

-0.72

arts

-0.70

arter

-0.69

atron

-0.68

estern

-0.67

regon

-0.65

amina

-0.65

obal

-0.64

POSITIVE LOGITS

.(

0.67

geist

0.67

 incarnation

0.63

NULL

0.63

mAh

0.62

 Aviv

0.61

âĢ²

0.60

dule

0.60

ãĤ©

0.60

OGR

0.60

Activations Density 0.014%

dates or time-related phrases

references to a specific temporal context or timeframe

No Comments

No Known Activations