INDEX

Explanations

words related to government and political entities

oai_token-act-pair · gpt-3.5-turbo

references to governmental and legal entities or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

hots

-0.74

hops

-0.72

ometimes

-0.65

states

-0.64

icles

-0.63

lehem

-0.61

utenant

-0.61

af

-0.60

offic

-0.60

married

-0.60

POSITIVE LOGITS

's

1.19

 takeover

1.00

 version

0.90

ÃŃs

0.85

 approach

0.83

 proposal

0.82

 plan

0.82

 mandate

0.80

 requirement

0.80

 shutdown

0.80

Activations Density 0.264%

words related to government and political entities

references to governmental and legal entities or actions

No Comments

No Known Activations