INDEX

Explanations

phrases related to minimizing something, such as minimizing risk or damage

oai_token-act-pair · gpt-3.5-turbo

concepts related to reduction or minimization of negative impacts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ker

-0.87

king

-0.82

alyst

-0.76

ansas

-0.74

intel

-0.69

enegger

-0.69

kers

-0.67

soon

-0.66

otle

-0.63

piring

-0.63

POSITIVE LOGITS

 minimize

0.91

imize

0.89

 minimizing

0.82

 minim

0.82

 distractions

0.74

 minimized

0.73

urized

0.72

 maximize

0.71

 misunder

0.70

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.70

Activations Density 0.012%

phrases related to minimizing something, such as minimizing risk or damage

concepts related to reduction or minimization of negative impacts

No Comments

No Known Activations