INDEX

Explanations

references to data input and processing

oai_token-act-pair · gpt-3.5-turbo

references to input or data submitted within a technical context

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

 Seraph

-0.70

llan

-0.64

 Crusade

-0.63

astern

-0.63

 Wright

-0.62

neys

-0.62

esa

-0.61

ulet

-0.61

itect

-0.59

ISS

-0.59

POSITIVE LOGITS

ted

1.10

 inputs

0.91

ting

0.89

ters

0.88

 input

0.85

enza

0.83

mitter

0.74

ifferent

0.74

prise

0.74

ierrez

0.73

Activations Density 0.011%

references to data input and processing

references to input or data submitted within a technical context

No Comments

No Known Activations