INDEX

Explanations

phrases related to an intense pursuit or search

oai_token-act-pair · gpt-3.5-turbo

references to "hunts" or hunting-related activities

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

 Consent

-0.69

 Kissinger

-0.68

Dys

-0.67

medi

-0.66

 Disability

-0.66

 Equity

-0.63

anes

-0.62

 Static

-0.60

 Sachs

-0.60

 Errors

-0.59

POSITIVE LOGITS

 hunt

3.81

 hunts

2.66

hunt

2.39

 hunted

2.23

 Hunt

2.04

 hunting

2.02

Hunt

1.96

 hunters

1.85

 hunter

1.72

 chase

1.68

Activations Density 0.012%

phrases related to an intense pursuit or search

references to "hunts" or hunting-related activities

No Comments

No Known Activations