INDEX

Explanations

words related to protection or security

oai_token-act-pair · gpt-3.5-turbo

phrases related to the act of safeguarding or defense

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

jet

-0.74

Discussion

-0.73

sb

-0.72

dl

-0.70

hour

-0.70

mpeg

-0.70

zig

-0.69

ppa

-0.67

translation

-0.67

RT

-0.67

POSITIVE LOGITS

 protecting

1.23

 shielding

1.06

 protects

1.00

 safegu

0.97

 adolesc

0.94

 preserving

0.90

 preventing

0.88

 shielded

0.87

 guarding

0.86

 conduc

0.85

Activations Density 0.016%

words related to protection or security

phrases related to the act of safeguarding or defense

No Comments

No Known Activations