INDEX

Explanations

mentions of blades

oai_token-act-pair · gpt-3.5-turbo

references to blades, particularly in terms of their features or context in weapons and tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

bos

-0.80

gdala

-0.79

ADRA

-0.78

upon

-0.73

riad

-0.72

ally

-0.72

quickShipAvailable

-0.71

tics

-0.69

ordable

-0.69

obyl

-0.66

POSITIVE LOGITS

 blades

1.26

 blade

1.12

tips

1.01

blade

0.97

 Runner

0.96

 Cutter

0.92

Blade

0.91

 Blade

0.89

 knives

0.89

 Blades

0.86

Activations Density 0.009%

mentions of blades

references to blades, particularly in terms of their features or context in weapons and tools

No Comments

No Known Activations