INDEX

Explanations

phrases related to government entities and procedures, specifically the "State Department."

oai_token-act-pair · gpt-3.5-turbo

mentions of the "State Department" or references related to state governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uckland

-0.78

irsch

-0.76

peria

-0.75

elled

-0.74

azi

-0.74

ussen

-0.73

emis

-0.72

 Aires

-0.71

undai

-0.71

ateurs

-0.71

POSITIVE LOGITS

 State

1.05

wide

1.04

 Governments

0.91

 Dept

0.87

 Correctional

0.84

 Legisl

0.83

State

0.82

 Alchemist

0.80

 Legislature

0.79

 States

0.79

Activations Density 0.023%

phrases related to government entities and procedures, specifically the "State Department."

mentions of the "State Department" or references related to state governance

No Comments

No Known Activations