INDEX

Explanations

references to locations within conflict zones related to Syria

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "Syrian" in various contexts

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

ertodd

-0.90

inarily

-0.87

otle

-0.85

owler

-0.82

print

-0.79

ocobo

-0.78

WARD

-0.78

inburgh

-0.78

ridges

-0.76

ATA

-0.76

POSITIVE LOGITS

 refugees

1.11

 Arab

1.10

 Refugees

1.05

 refugee

1.03

 Kurds

0.97

 Syrians

0.91

 Syrian

0.89

 Refugee

0.88

 migrants

0.87

Syrian

0.86

Activations Density 0.012%

references to locations within conflict zones related to Syria

mentions of the term "Syrian" in various contexts

No Comments

No Known Activations