INDEX

Explanations

terms related to political discussions around immigration

oai_token-act-pair · gpt-3.5-turbo

mentions of immigration policy and reform discussions

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

ilts

-0.83

MpServer

-0.82

eenth

-0.80

oeuv

-0.77

sonian

-0.76

å§«

-0.75

âĶĢâĶĢ

-0.75

olls

-0.74

Ò

-0.73

IVES

-0.73

POSITIVE LOGITS

 immigration

1.05

 detention

1.04

 visas

0.96

 deportation

0.89

 reform

0.87

 Detention

0.86

 amnesty

0.86

 deported

0.86

 immigrants

0.86

 Immigration

0.84

Activations Density 0.025%

terms related to political discussions around immigration

mentions of immigration policy and reform discussions

No Comments

No Known Activations