INDEX

Explanations

phrases related to political and social issues

oai_token-act-pair · gpt-3.5-turbo

the word "and" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ishi

-0.72

Deal

-0.71

bo

-0.68

iors

-0.67

elle

-0.67

anas

-0.66

ire

-0.64

Beck

-0.64

zzle

-0.64

OTA

-0.64

POSITIVE LOGITS

 consequently

0.86

romeda

0.85

 therefore

0.84

 other

0.83

 assorted

0.78

 vice

0.74

rogen

0.73

rogens

0.69

 others

0.68

 preservation

0.67

Activations Density 0.459%

phrases related to political and social issues

the word "and" in various contexts

No Comments

No Known Activations