INDEX

Explanations

words related to linguistic and morphological analysis

oai_token-act-pair · gpt-3.5-turbo

words and terms related to scientific and academic disciplines

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

 Faster

-0.77

KEN

-0.73

ortium

-0.73

lain

-0.72

lua

-0.70

ãģ®éŃĶ

-0.69

UV

-0.69

IRD

-0.69

 Leilan

-0.65

 âĶľ

-0.65

POSITIVE LOGITS

ities

0.98

ity

0.97

 properties

0.92

 warfare

0.86

 attainment

0.83

itÃ©

0.81

ALLY

0.81

 superiority

0.80

 prowess

0.80

 qualities

0.80

Activations Density 0.082%

words related to linguistic and morphological analysis

words and terms related to scientific and academic disciplines

No Comments

No Known Activations