INDEX

Explanations

forms of the word "Making" at varying activation levels

oai_token-act-pair · gpt-3.5-turbo

instances of the word "making" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Mub

-0.70

 Vest

-0.69

ban

-0.67

---------

-0.67

 scrimmage

-0.64

 nurs

-0.61

 vest

-0.58

TERN

-0.58

 conditioning

-0.58

-0.57

POSITIVE LOGITS

 sure

0.88

hift

0.87

ÄŁ

0.86

awaru

0.83

itives

0.82

emort

0.82

akable

0.78

URE

0.75

amera

0.73

ailable

0.73

Activations Density 0.034%

forms of the word "Making" at varying activation levels

instances of the word "making" in various contexts

No Comments

No Known Activations