INDEX

Explanations

phrases related to decision-making and policies

oai_token-act-pair · gpt-3.5-turbo

phrases related to decision-making processes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ITED

-0.68

ModLoader

-0.67

 Rusty

-0.63

irl

-0.63

uously

-0.62

 accredited

-0.61

 Veter

-0.61

estial

-0.61

 havoc

-0.60

iless

-0.59

POSITIVE LOGITS

making

1.57

makers

1.25

taking

1.12

maker

1.09

telling

1.06

process

1.05

processing

1.04

tree

0.99

Making

0.99

thinking

0.96

Activations Density 0.047%

phrases related to decision-making and policies

phrases related to decision-making processes

No Comments

No Known Activations