INDEX

Explanations

numbers or measurements in the format of "ths"

oai_token-act-pair · gpt-3.5-turbo

specific quantities or measurements, particularly related to "ths" (tenths or thousands)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Byrne

-0.67

 patronage

-0.65

DOI

-0.62

edited

-0.62

ITED

-0.61

 multim

-0.59

 Marriott

-0.57

 solicitor

-0.57

 mileage

-0.56

EMENT

-0.56

POSITIVE LOGITS

ths

1.05

ulhu

0.95

ippi

0.88

ystem

0.88

ttp

0.87

mith

0.85

linger

0.85

terday

0.85

lap

0.84

tml

0.84

Activations Density 0.009%

numbers or measurements in the format of "ths"

specific quantities or measurements, particularly related to "ths" (tenths or thousands)

No Comments

No Known Activations