INDEX

Explanations

terms related to authority and government

oai_token-act-pair · gpt-3.5-turbo

terms related to governance and authority

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

umo

-0.78

sonian

-0.70

ertodd

-0.70

uder

-0.68

uni

-0.67

orescent

-0.65

ewitness

-0.65

imus

-0.65

Opening

-0.64

eele

-0.63

POSITIVE LOGITS

 governing

1.17

 governance

0.97

 govern

0.85

 affairs

0.81

 structure

0.80

 structures

0.80

 Govern

0.77

 Reform

0.76

eering

0.74

 regimes

0.74

Activations Density 0.019%

terms related to authority and government

terms related to governance and authority

No Comments

No Known Activations