INDEX

Explanations

numerical values and company names

oai_token-act-pair · gpt-3.5-turbo

numerical data and special characters, likely related to statistics or measurements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

="/

-0.67

 [âĢ¦]

-0.66

ONSORED

-0.63

.–

-0.63

atform

-0.62

 inherently

-0.58

":["

-0.56

-[

-0.55

âĤ¬

-0.55

 Tues

-0.54

POSITIVE LOGITS

olate

0.81

aneous

0.80

artifacts

0.75

ronic

0.73

ibaba

0.73

BuyableInstoreAndOnline

0.70

antage

0.68

ortium

0.68

ifling

0.67

uncle

0.67

Activations Density 0.164%

numerical values and company names

numerical data and special characters, likely related to statistics or measurements

No Comments

No Known Activations