INDEX

Explanations

phrases or terms related to the abbreviation "MP"

oai_token-act-pair · gpt-3.5-turbo

references to the term "MP," likely indicating a focus on members of parliament or similar political figures

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

xual

-0.99

stal

-0.79

renheit

-0.77

idad

-0.70

uran

-0.70

hook

-0.68

color

-0.68

amus

-0.66

gender

-0.65

kefeller

-0.65

POSITIVE LOGITS

MP

1.20

MP

0.95

PLIC

0.92

MLA

0.87

rint

0.86

LAB

0.85

mp

0.83

JP

0.81

FS

0.76

AA

0.76

Activations Density 0.007%

phrases or terms related to the abbreviation "MP"

references to the term "MP," likely indicating a focus on members of parliament or similar political figures

No Comments

No Known Activations