INDEX

Explanations

technology-related terms, specifically related to technology companies like Lenovo and Motorola

oai_token-act-pair · gpt-3.5-turbo

mentions of specific technology brands

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

outh

-0.71

ippi

-0.67

 drag

-0.67

jam

-0.65

sob

-0.65

åº

-0.64

ocaust

-0.64

hop

-0.62

Cup

-0.62

 fucked

-0.61

POSITIVE LOGITS

 Lenovo

2.95

 Motorola

2.59

 Panasonic

2.32

 VMware

2.20

 Nokia

2.17

 Dell

2.16

 Huawei

2.08

 Cisco

2.07

IBM

2.06

 Xiaomi

1.94

Activations Density 0.050%

technology-related terms, specifically related to technology companies like Lenovo and Motorola

mentions of specific technology brands

No Comments

No Known Activations