INDEX

Explanations

conditional statements with "if" and "So" indicating logical relationships

oai_token-act-pair · gpt-3.5-turbo

phrases that begin with "So" or similar conjunctions that introduce explanations or clarifications

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

Not in Any Lists

No Comments

Negative Logits

intosh

-0.68

Â¢

-0.63

ropolitan

-0.61

ï¸

-0.60

gall

-0.60

-0.59

 Written

-0.59

Truth

-0.58

exclusive

-0.58

kb

-0.58

POSITIVE LOGITS

 theoretically

1.04

oner

0.91

 ideally

0.87

 beware

0.82

fter

0.79

 technically

0.79

 whoever

0.78

 unless

0.76

oths

0.75

ivably

0.74

Activations Density 0.118%

conditional statements with "if" and "So" indicating logical relationships

phrases that begin with "So" or similar conjunctions that introduce explanations or clarifications

No Comments

No Known Activations