INDEX

Explanations

phrases related to governmental actions and policies

oai_token-act-pair · gpt-3.5-turbo

mentions of government administrations and their actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

die

-0.73

 Spect

-0.72

 Circle

-0.72

die

-0.71

 dies

-0.69

 dying

-0.67

 Dying

-0.66

 Slayer

-0.66

Die

-0.66

 compat

-0.65

POSITIVE LOGITS

 policies

0.89

 appoint

0.89

 unveil

0.89

 deliberations

0.87

 bung

0.86

 announce

0.84

 declass

0.84

 deliber

0.82

 overhaul

0.82

 promulg

0.81

Activations Density 0.527%

phrases related to governmental actions and policies

mentions of government administrations and their actions

No Comments

No Known Activations