INDEX

Explanations

names of characters from various shows and movies

oai_token-act-pair · gpt-3.5-turbo

words with frequent vowel usage, particularly those ending in 'a', 'ie', and 'ei'

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

ounding

-0.82

itutional

-0.78

oter

-0.77

krit

-0.75

orer

-0.75

rador

-0.74

ebin

-0.73

ensibly

-0.72

igm

-0.72

orage

-0.72

POSITIVE LOGITS

 Spears

0.91

 Rivera

0.90

 Lann

0.88

 Gomez

0.88

 Kardashian

0.87

Rae

0.87

 Jenner

0.84

 Hath

0.83

Pry

0.83

 Torres

0.82

Activations Density 0.181%

names of characters from various shows and movies

words with frequent vowel usage, particularly those ending in 'a', 'ie', and 'ei'

No Comments

No Known Activations