INDEX

Explanations

phrases related to policies or decisions that have significant impact or controversy

oai_token-act-pair · gpt-3.5-turbo

topics related to controversial policies and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 ALWAYS

-0.83

 murdered

-0.81

 ignor

-0.79

 fucking

-0.78

 NEVER

-0.77

 murdering

-0.76

 randomly

-0.75

 pathological

-0.70

 raping

-0.70

 âī¥

-0.69

POSITIVE LOGITS

 cheered

0.89

 optimism

0.81

 Macron

0.78

 upbeat

0.77

plom

0.76

 Osborne

0.76

 Grassley

0.75

 boon

0.75

 setback

0.75

 welcomed

0.74

Activations Density 0.464%

phrases related to policies or decisions that have significant impact or controversy

topics related to controversial policies and their implications

No Comments

No Known Activations