INDEX

Explanations

technical terms related to procedures or techniques

oai_token-act-pair · gpt-3.5-turbo

references to various methods or techniques

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

brance

-0.71

uph

-0.70

cone

-0.69

reath

-0.67

ikk

-0.67

hare

-0.66

obe

-0.66

sey

-0.66

ominium

-0.66

osphere

-0.65

POSITIVE LOGITS

 methods

3.61

 Methods

2.72

 method

2.43

 techniques

2.42

method

2.38

Methods

2.26

 Method

1.96

Method

1.93

 procedures

1.82

 tactics

1.79

Activations Density 0.014%

technical terms related to procedures or techniques

references to various methods or techniques

No Comments

No Known Activations