INDEX

Explanations

spelling of specific acronyms or identifiers related to political or governmental concepts

oai_token-act-pair · gpt-3.5-turbo

acronyms or abbreviations related to specific organizations, assessments, or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

acity

-0.71

things

-0.64

 marvelous

-0.61

 unpop

-0.59

rooms

-0.58

 Allaah

-0.58

 ç¥ŀ

-0.58

taboola

-0.57

metadata

-0.57

esville

-0.56

POSITIVE LOGITS

Bs

1.16

Cs

1.14

Os

1.11

FU

1.10

MO

1.10

Ds

1.09

Ts

1.07

Ps

1.05

1.04

Gs

1.04

Activations Density 0.258%

spelling of specific acronyms or identifiers related to political or governmental concepts

acronyms or abbreviations related to specific organizations, assessments, or processes

No Comments

No Known Activations