INDEX

Explanations

mentions of sharp objects or blades, potentially related to weaponry

oai_token-act-pair · gpt-3.5-turbo

references to blades, particularly in various contexts, including knives and tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

quickShipAvailable

-0.89

ADRA

-0.80

gdala

-0.80

bos

-0.79

 Niet

-0.76

ESA

-0.71

ravel

-0.70

riad

-0.70

phis

-0.68

EMBER

-0.67

POSITIVE LOGITS

 blades

1.12

tips

1.03

 Runner

0.98

 blade

0.97

tip

0.90

blade

0.89

Blade

0.83

 Cutter

0.82

 knives

0.80

hawk

0.80

Activations Density 0.019%

mentions of sharp objects or blades, potentially related to weaponry

references to blades, particularly in various contexts, including knives and tools

No Comments

No Known Activations