INDEX

Explanations

mentions of specific proper nouns or entities

oai_token-act-pair · gpt-3.5-turbo

vocabulary related to criminal justice and incarceration

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

 thereof

-0.79

 ..."

-0.78

-0.75

 thereto

-0.74

 âĢ¦"

-0.73

 respectively

-0.71

 è£ı

-0.68

..."

-0.63

#$

-0.62

ãĤ¼ãĤ¦ãĤ¹

-0.61

POSITIVE LOGITS

osate

0.80

 Profile

0.76

 Advice

0.71

 Oversight

0.67

 Overview

0.67

 Analysis

0.67

 Benefits

0.67

 Appeal

0.67

 Accountability

0.66

pedia

0.65

Activations Density 0.317%

mentions of specific proper nouns or entities

vocabulary related to criminal justice and incarceration

No Comments

No Known Activations