INDEX

Explanations

terms related to governmental and bureaucratic entities, particularly those with the word "Federal" in them

oai_token-act-pair · gpt-3.5-turbo

references to the Federal government and its agencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lihood

-0.82

paces

-0.76

onge

-0.75

IED

-0.73

enger

-0.71

 Carbuncle

-0.71

oldemort

-0.70

pered

-0.68

Kik

-0.68

ammers

-0.67

POSITIVE LOGITS

 Reserve

1.08

ist

0.99

ists

0.98

ism

0.89

 Aviation

0.84

istic

0.84

 Election

0.79

tarian

0.77

 Bureau

0.77

 Government

0.77

Activations Density 0.023%

terms related to governmental and bureaucratic entities, particularly those with the word "Federal" in them

references to the Federal government and its agencies

No Comments

No Known Activations