INDEX

Explanations

terms related to government actions or legal matters, specifically those involving federal entities

oai_token-act-pair · gpt-3.5-turbo

references to federal government 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

zzo

-0.90

sbm

-0.82

enger

-0.80

 Clever

-0.80

lihood

-0.79

 Rooms

-0.79

 Tycoon

-0.78

bringer

-0.77

bane

-0.76

hops

-0.75

POSITIVE LOGITS

 government

1.06

 appellate

0.97

 agencies

0.91

 agency

0.88

 authorities

0.87

 judiciary

0.85

 institution

0.82

 administration

0.82

law

0.82

 oversight

0.81

Activations Density 0.026%

terms related to government actions or legal matters, specifically those involving federal entities

references to federal government entities or actions

No Comments

No Known Activations