INDEX

Explanations

terms related to governmental entities and activities

oai_token-act-pair · gpt-3.5-turbo

references to government and related activities or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

CONT

-0.82

Dialog

-0.72

Rank

-0.69

ous

-0.67

 Cous

-0.66

 Tome

-0.66

XVI

-0.66

 Schwar

-0.66

bring

-0.66

tom

-0.65

POSITIVE LOGITS

 agencies

0.97

 subsidized

0.96

 government

0.91

 mandated

0.89

 shutdown

0.88

 bureaucracy

0.87

 schooling

0.83

 officials

0.83

 agency

0.82

 administrator

0.80

Activations Density 0.063%

terms related to governmental entities and activities

references to government and related activities or actions

No Comments

No Known Activations