INDEX

Explanations

words related to laboratory work

oai_token-act-pair · gpt-3.5-turbo

references to a laboratory environment

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

theless

-0.82

 Chall

-0.67

iversal

-0.66

thood

-0.66

 Resolution

-0.64

lihood

-0.64

vironment

-0.62

 Masquerade

-0.61

 Newly

-0.60

itarian

-0.60

POSITIVE LOGITS

rador

1.41

elling

1.15

orer

1.15

lab

1.07

 labs

0.94

elled

0.92

yrinth

0.90

lab

0.90

rosse

0.89

orers

0.88

Activations Density 0.008%

words related to laboratory work

references to a laboratory environment

No Comments

No Known Activations