INDEX

Explanations

terms related to exploration or investigation

oai_token-act-pair · gpt-3.5-turbo

words related to exploration or investigating concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

maid

-0.70

 Nadu

-0.69

ndra

-0.68

Downloadha

-0.68

chery

-0.67

PRES

-0.66

manship

-0.66

interstitial

-0.65

pora

-0.64

 elves

-0.62

POSITIVE LOGITS

oded

1.13

oding

1.07

icit

1.05

oit

1.04

ained

1.00

osion

0.99

odes

0.97

icably

0.96

orer

0.95

orers

0.94

Activations Density 0.010%

terms related to exploration or investigation

words related to exploration or investigating concepts

No Comments

No Known Activations