INDEX

Explanations

phrases indicating a subjective evaluation or perception of a situation or idea

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate perception or subjective judgment about various subjects

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

aughs

-0.71

mers

-0.68

urrence

-0.67

anwhile

-0.64

rowing

-0.63

cised

-0.62

isin

-0.60

ioch

-0.60

 Strikes

-0.60

 Hunts

-0.60

POSITIVE LOGITS

 innocuous

1.05

 daunting

1.00

 intimidating

0.93

 intuitive

0.92

 confusing

0.90

 insignificant

0.89

intuitive

0.87

 contradictory

0.86

 trivial

0.86

 superf

0.85

Activations Density 0.064%

phrases indicating a subjective evaluation or perception of a situation or idea

phrases that indicate perception or subjective judgment about various subjects

No Comments

No Known Activations