INDEX

Explanations

names of political figures

oai_token-act-pair · gpt-3.5-turbo

mentions of prominent political figures, particularly Narendra Modi

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hops

-0.74

ascript

-0.72

erie

-0.68

CAP

-0.68

inventoryQuantity

-0.68

Transfer

-0.68

itone

-0.68

hor

-0.67

oning

-0.66

Haunted

-0.66

POSITIVE LOGITS

 Modi

1.20

 Narendra

0.89

 Tayyip

0.86

 Kumar

0.86

 Hussein

0.84

 Kejriwal

0.72

 Pradesh

0.71

 Duterte

0.70

 Trudeau

0.68

isance

0.68

Activations Density 0.014%

names of political figures

mentions of prominent political figures, particularly Narendra Modi

No Comments

No Known Activations