INDEX

Explanations

financial terms or acronyms related to businesses or organizations

oai_token-act-pair · gpt-3.5-turbo

the instances of the term "OM" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wall

-0.76

 whatever

-0.74

hs

-0.71

place

-0.67

 otherwise

-0.66

reality

-0.63

waters

-0.62

Fen

-0.62

real

-0.62

ãĥĨãĤ£

-0.61

POSITIVE LOGITS

OM

1.23

orrow

1.01

FTWARE

1.00

obile

0.99

EGA

0.98

OGR

0.98

edia

0.96

ovember

0.95

ODE

0.93

atis

0.92

Activations Density 0.004%

financial terms or acronyms related to businesses or organizations

the instances of the term "OM" across various contexts

No Comments

No Known Activations