INDEX

Explanations

words related to change or alteration

oai_token-act-pair · gpt-3.5-turbo

terms related to changes and adjustments in systems or strategies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

putable

-0.71

otal

-0.68

 Piercing

-0.67

ymph

-0.65

ierce

-0.64

GV

-0.63

 Peak

-0.62

riot

-0.61

è¦ļéĨĴ

-0.60

 Mechdragon

-0.60

POSITIVE LOGITS

 accordingly

1.13

 fortunes

0.87

 drastically

0.84

 dramatically

0.81

 habits

0.79

 differently

0.72

 radically

0.70

 altogether

0.68

 dial

0.68

 hinges

0.67

Activations Density 0.447%

words related to change or alteration

terms related to changes and adjustments in systems or strategies

No Comments

No Known Activations