INDEX

Explanations

references to specific titles, most likely related to movies, books or games

oai_token-act-pair · gpt-3.5-turbo

the word "the" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

arella

-0.83

arten

-0.78

BER

-0.67

aciously

-0.64

 Canaver

-0.64

 furthermore

-0.64

PsyNetMessage

-0.63

etsk

-0.63

 barr

-0.61

pload

-0.60

POSITIVE LOGITS

 aforementioned

0.96

 same

0.95

 smallest

0.95

 utmost

0.91

 latter

0.91

 greatest

0.83

 Americas

0.82

 highest

0.82

 largest

0.81

 Ancients

0.79

Activations Density 0.383%

references to specific titles, most likely related to movies, books or games

the word "the" in various contexts

No Comments

No Known Activations