INDEX

Explanations

words related to symbology and medical conditions

oai_token-act-pair · gpt-3.5-turbo

instances of the term "sy"

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

ãģ®éŃĶ

-0.70

ORED

-0.69

inarily

-0.66

ãĥ¼ãĥĨãĤ£

-0.66

ardless

-0.62

headers

-0.60

DERR

-0.60

 Miko

-0.59

ãģĤ

-0.59

ificate

-0.59

POSITIVE LOGITS

mbol

1.42

thia

1.23

ntax

1.22

racuse

1.06

nergy

1.04

nerg

1.03

philis

0.99

ptoms

0.97

bil

0.97

rup

0.91

Activations Density 0.021%

words related to symbology and medical conditions

instances of the term "sy"

No Comments

No Known Activations