INDEX

Explanations

phrases related to security breaches or vulnerabilities

oai_token-act-pair · gpt-3.5-turbo

terms related to security vulnerabilities and unsecured systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĥ£

-0.87

è¦ļéĨĴ

-0.83

lift

-0.77

 Forsaken

-0.74

lihood

-0.68

 Warcraft

-0.66

ãĥīãĥ©ãĤ´ãĥ³

-0.65

çİĭ

-0.64

ãģ¦

-0.64

 Hots

-0.64

POSITIVE LOGITS

rets

1.25

urities

1.09

enaries

1.03

utions

1.03

recy

0.99

rete

0.97

ular

0.96

aucus

0.94

RET

0.93

ugu

0.92

Activations Density 0.012%

phrases related to security breaches or vulnerabilities

terms related to security vulnerabilities and unsecured systems

No Comments

No Known Activations