INDEX

Explanations

repeated occurrences of the word "same"

oai_token-act-pair · gpt-3.5-turbo

expressions of similarity or sameness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

*=-

-0.72

omics

-0.69

 Provided

-0.64

orsi

-0.64

akings

-0.61

åĪ

-0.60

ommod

-0.60

arest

-0.60

OST

-0.59

erest

-0.59

POSITIVE LOGITS

 thing

0.96

 exact

0.91

 vein

0.88

 amount

0.81

rity

0.74

 sorts

0.71

 kind

0.70

day

0.70

 kinds

0.67

same

0.67

Activations Density 0.045%

repeated occurrences of the word "same"

expressions of similarity or sameness

No Comments

No Known Activations