INDEX

Explanations

terms related to trade or trading activities

oai_token-act-pair · gpt-3.5-turbo

references to trade activities or agreements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

isson

-0.74

imen

-0.67

odder

-0.66

acid

-0.66

mary

-0.66

indal

-0.63

 Rita

-0.62

rawn

-0.62

 Adren

-0.61

 Reflex

-0.61

POSITIVE LOGITS

 trade

3.84

trade

2.95

 Trade

2.80

Trade

2.70

 trades

2.57

 trading

2.18

 traded

1.98

 Trading

1.64

 traders

1.63

 trader

1.57

Activations Density 0.013%

terms related to trade or trading activities

references to trade activities or agreements

No Comments

No Known Activations