INDEX

Explanations

phrases relating to laws, regulations, and political terms

oai_token-act-pair · gpt-3.5-turbo

terms related to legal and political issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

alks

-0.96

lees

-0.94

ravings

-0.94

assies

-0.90

abases

-0.88

mails

-0.88

tags

-0.88

 ceilings

-0.84

ouses

-0.84

ramids

-0.83

POSITIVE LOGITS

 affair

1.32

 phenomenon

1.17

 manifestation

1.12

 reminder

1.11

 conduit

1.11

 cornerstone

1.09

 nightmare

1.07

 fixture

1.05

 staple

1.05

 indicator

1.05

Activations Density 0.491%

phrases relating to laws, regulations, and political terms

terms related to legal and political issues

No Comments

No Known Activations