INDEX

Explanations

mentions of political administrations, specifically focusing on the Trump administration

oai_token-act-pair · gpt-3.5-turbo

references to the Trump administration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ILCS

-0.72

 Mata

-0.67

Ô

-0.66

#$#$

-0.66

chens

-0.65

 Plate

-0.63

ãĤ½

-0.63

cible

-0.62

ãĤ¡

-0.62

AUT

-0.61

POSITIVE LOGITS

 officials

0.92

 secretaries

0.81

 appoint

0.78

 helicop

0.75

 playbook

0.74

 nominees

0.74

 official

0.74

 policies

0.74

 administration

0.74

 shutdown

0.73

Activations Density 0.042%

mentions of political administrations, specifically focusing on the Trump administration

references to the Trump administration

No Comments

No Known Activations