INDEX

Explanations

phrases separated by asterisks

oai_token-act-pair · gpt-3.5-turbo

repeated asterisks or similar formatting indicators that could imply emphasis or censorship

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

etheless

-0.76

 conservation

-0.74

 erosion

-0.73

vation

-0.73

ugal

-0.71

 migr

-0.71

srf

-0.69

oidal

-0.67

 scattering

-0.67

 radi

-0.67

POSITIVE LOGITS

***

0.97

***

0.94

-->

0.87

=-=-

0.86

-+

0.81

IER

0.80

!/

0.79

>>>

0.77

~~~~

0.76

REDACTED

0.75

Activations Density 0.012%

phrases separated by asterisks

repeated asterisks or similar formatting indicators that could imply emphasis or censorship

No Comments

No Known Activations