INDEX

Explanations

topics related to national security

oai_token-act-pair · gpt-3.5-turbo

terms related to national security

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

chrom

-0.83

amaz

-0.79

sample

-0.78

ãĥĺ

-0.77

 Refresh

-0.76

ken

-0.76

oven

-0.76

RGB

-0.76

Syn

-0.74

 Niet

-0.73

POSITIVE LOGITS

 advisor

1.09

 adviser

1.08

 Advisor

0.96

 Adviser

0.91

 planners

0.87

 strategist

0.86

olitics

0.84

 emergencies

0.84

 apparatus

0.83

policy

0.82

Activations Density 0.023%

topics related to national security

terms related to national security

No Comments

No Known Activations