INDEX

Explanations

statements expressing opinions or beliefs

oai_token-act-pair · gpt-3.5-turbo

assertions and claims about opinions or beliefs

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

hander

-0.71

ritten

-0.68

phrine

-0.68

ãĤ´ãĥ³

-0.67

atri

-0.67

bid

-0.67

written

-0.67

recorded

-0.65

pin

-0.64

 scanned

-0.63

POSITIVE LOGITS

 otherwise

1.18

 disrespect

0.87

 excuses

0.79

 anything

0.76

 baseless

0.73

 unreasonable

0.70

 incorrectly

0.69

 dismissing

0.68

 neglect

0.68

 criticizing

0.67

Activations Density 0.271%

statements expressing opinions or beliefs

assertions and claims about opinions or beliefs

No Known Activations