INDEX

Explanations

phrases related to technology and numerical data

oai_token-act-pair · gpt-3.5-turbo

specific terms related to statistics, trends, and organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

depending

-0.86

assian

-0.74

ablishment

-0.68

necessary

-0.66

any

-0.65

inational

-0.65

 describ

-0.64

onymous

-0.64

 conflic

-0.63

universal

-0.62

POSITIVE LOGITS

MFT

0.68

 Decoder

0.64

Ltd

0.64

LTD

0.64

 âĵĺ

0.62

ensis

0.62

 Edition

0.62

Pt

0.61

çīĪ

0.59

 Sniper

0.59

Activations Density 0.675%

phrases related to technology and numerical data

specific terms related to statistics, trends, and organizational structures

No Comments

No Known Activations