INDEX

Explanations

phrases related to interpreting or analyzing information

oai_token-act-pair · gpt-3.5-turbo

phrases related to interpretation or understanding of legal concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

EStreamFrame

-0.82

 slew

-0.79

utenberg

-0.67

authorized

-0.65

 dispatched

-0.65

 authorised

-0.65

 responsibly

-0.63

 sqor

-0.62

 Marketable

-0.60

ensibly

-0.60

POSITIVE LOGITS

 Dill

0.71

 Clarks

0.68

bb

0.65

icho

0.64

-+-+

0.63

ippi

0.63

Pak

0.61

 Chero

0.61

 Boards

0.60

cho

0.60

Activations Density 0.000%

phrases related to interpreting or analyzing information

phrases related to interpretation or understanding of legal concepts

No Comments

No Known Activations