INDEX

Explanations

occurrences of the prefix "un-" in words

oai_token-act-pair · gpt-4o-mini Triggered by @bot

proper nouns related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

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

 tomat

-0.71

Peb

-0.70

Ö¼

-0.67

 Madden

-0.67

 accur

-0.65

ãĤ¼

-0.65

mos

-0.63

 houses

-0.62

 improv

-0.60

ortment

-0.59

POSITIVE LOGITS

geon

1.23

geons

1.19

nel

1.18

nels

1.05

fortunately

1.04

ned

1.04

ners

1.03

cles

1.01

iversity

1.00

culus

0.99

Activations Density 0.019%

occurrences of the prefix "un-" in words

proper nouns related to geographical locations

No Comments

No Known Activations