INDEX

Explanations

words related to alphabets and symbols

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific character or symbol

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WARD

-0.71

 Coliseum

-0.69

waves

-0.69

 commute

-0.68

wards

-0.67

ITNESS

-0.66

 rush

-0.65

 microw

-0.64

iflower

-0.64

 Neural

-0.64

POSITIVE LOGITS

Å

1.41

¼

1.32

½

1.23

ı

1.18

ĭ

1.12

ĵ

1.12

¾

1.12

ł

1.09

ĥ

1.09

»

1.09

Activations Density 0.007%

words related to alphabets and symbols

occurrences of a specific character or symbol

No Comments

No Known Activations