INDEX

Explanations

phrases related to business processes and monitoring

oai_token-act-pair · gpt-3.5-turbo

terms related to processes and monitoring in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 ingest

-0.82

 rampant

-0.81

 lethal

-0.80

 breaching

-0.75

 habitual

-0.75

 wealthy

-0.75

 skilled

-0.74

 ancestral

-0.71

 exile

-0.70

 upset

-0.70

POSITIVE LOGITS

 Begins

1.29

 Starts

1.23

 Costs

1.20

 Overview

1.20

 Coverage

1.19

 Areas

1.19

 Functions

1.19

 Profile

1.16

 Shows

1.16

 Comparison

1.16

Activations Density 0.330%

phrases related to business processes and monitoring

terms related to processes and monitoring in various contexts

No Comments

No Known Activations