INDEX

Explanations

commands or actions related to execution

oai_token-act-pair · gpt-3.5-turbo

terms related to the action of executing tasks or commands

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

 Soda

-0.66

 Greater

-0.66

 Tire

-0.65

 Ness

-0.65

 Mens

-0.64

conom

-0.64

 Kens

-0.63

âĸł

-0.61

vol

-0.61

 Lank

-0.60

POSITIVE LOGITS

 execute

1.15

 executes

1.11

 executing

1.07

eering

1.00

 executed

0.99

 execution

0.98

 executions

0.96

[|

0.86

 Exec

0.82

ascript

0.82

Activations Density 0.013%

commands or actions related to execution

terms related to the action of executing tasks or commands

No Comments

No Known Activations