INDEX

Explanations

words related to being under supervision or being supervised

oai_token-act-pair · gpt-3.5-turbo

concepts related to supervision and oversight

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

grave

-0.80

bis

-0.76

govtrack

-0.70

ml

-0.67

UES

-0.67

false

-0.66

ãĥĥãĥī

-0.65

tr

-0.65

bp

-0.64

Sov

-0.63

POSITIVE LOGITS

 supervision

1.00

 supervised

0.87

 Instruct

0.80

 probation

0.79

 oversee

0.72

 superv

0.71

 conduc

0.71

vised

0.69

icka

0.67

 confir

0.67

Activations Density 0.031%

words related to being under supervision or being supervised

concepts related to supervision and oversight

No Comments

No Known Activations