INDEX

Explanations

proper nouns related to a specific character named "Nemo" or variations thereof

oai_token-act-pair · gpt-3.5-turbo

references to the name "Nem" or variations thereof 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

========

-0.74

============

-0.70

angers

-0.66

CONCLUS

-0.65

WAR

-0.65

ÃĽÃĽ

-0.65

 rehearsal

-0.65

======

-0.65

================================================================

-0.64

shapeshifter

-0.63

POSITIVE LOGITS

Nem

1.11

nem

1.09

esis

1.06

ominated

0.98

eny

0.92

ecast

0.92

 unden

0.89

elong

0.88

daq

0.86

acles

0.85

Activations Density 0.007%

proper nouns related to a specific character named "Nemo" or variations thereof

references to the name "Nem" or variations thereof in various contexts

No Comments

No Known Activations