INDEX

Explanations

terms related to intelligence and consequences

oai_token-act-pair · gpt-3.5-turbo

terms related to intelligence and consequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 shroud

-0.75

 hammer

-0.75

 viewing

-0.70

 sliding

-0.68

 crossing

-0.67

 Kaiser

-0.66

 walk

-0.66

 Petersen

-0.66

 attachment

-0.66

 funnel

-0.66

POSITIVE LOGITS

ential

1.40

ently

1.34

ences

1.33

entials

1.26

entially

1.26

ual

1.21

ably

1.17

ency

1.16

orial

1.16

ional

1.12

Activations Density 0.037%

terms related to intelligence and consequences

terms related to intelligence and consequence

No Comments

No Known Activations