INDEX

Explanations

proper nouns or titles of people or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "photo" 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 @ 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

sheet

-0.70

acious

-0.68

vertisement

-0.65

nesses

-0.64

innon

-0.64

ruary

-0.64

theless

-0.63

rity

-0.62

 occurrence

-0.61

 saline

-0.61

POSITIVE LOGITS

zzi

1.29

zzo

0.89

ÄŁ

0.80

oga

0.79

Ã±a

0.79

GP

0.79

onga

0.77

oto

0.76

redo

0.76

veland

0.75

Activations Density 0.018%

proper nouns or titles of people or events

occurrences of the word "photo" in various contexts

No Comments

No Known Activations