INDEX

Explanations

instances of the word "Iñigo" with varying activation levels, possibly indicating a focus on a specific person's name

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "igo."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

neapolis

-0.73

topic

-0.72

sheet

-0.72

ensional

-0.70

together

-0.68

abase

-0.66

practice

-0.65

sheets

-0.63

hetical

-0.62

itudes

-0.62

POSITIVE LOGITS

vernment

1.37

igo

1.01

 Libre

0.88

lio

0.84

ption

0.83

cia

0.80

cean

0.79

 Indigo

0.78

rosso

0.77

Viz

0.76

Activations Density 0.007%

instances of the word "Iñigo" with varying activation levels, possibly indicating a focus on a specific person's name

occurrences of the term "igo."

No Comments

No Known Activations