INDEX

Explanations

references to different offices, departments, and positions within an organizational structure

oai_token-act-pair · gpt-3.5-turbo

references to government offices and agencies

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

ashtra

-1.07

 verses

-0.80

ibling

-0.75

 passages

-0.72

ettel

-0.71

ivari

-0.71

estones

-0.71

 poems

-0.70

 breeds

-0.68

tarians

-0.67

POSITIVE LOGITS

 Strategic

0.93

 Budget

0.81

 Management

0.79

ãĤ±

0.79

 Integrity

0.76

 Personnel

0.76

 Inquiry

0.75

Correction

0.75

ãĥ¤

0.74

 Inspector

0.74

Activations Density 0.060%

references to different offices, departments, and positions within an organizational structure

references to government offices and agencies

No Comments

No Known Activations