INDEX

Explanations

acronyms ending in 'OS'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "OS" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ãĥ©ãĥ³

-0.85

ered

-0.77

ãĥ¬

-0.76

aneous

-0.72

ruary

-0.72

ries

-0.72

ãĥ¼ãĥĨ

-0.69

starter

-0.68

ãĤ©

-0.67

htt

-0.66

POSITIVE LOGITS

 ANGEL

1.03

OS

1.01

ECTION

0.94

DAQ

0.94

ystem

0.94

leep

0.90

omal

0.89

SI

0.88

DEM

0.86

pace

0.85

Activations Density 0.008%

acronyms ending in 'OS'

occurrences of the term "OS" and its variations

No Comments

No Known Activations