INDEX

Explanations

steps or instructions in a process

oai_token-act-pair · gpt-3.5-turbo

references to measurements and instructions in a detailed process

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

 Gund

-0.77

ORPG

-0.74

CVE

-0.72

asma

-0.67

ulence

-0.65

Moj

-0.64

UCT

-0.64

authent

-0.64

 Highly

-0.63

izable

-0.62

POSITIVE LOGITS

sts

1.01

 alternating

0.97

dc

0.96

Row

0.94

dc

0.93

 Repeat

0.91

 spaced

0.86

 repeat

0.85

row

0.85

 needle

0.84

Activations Density 0.124%

steps or instructions in a process

references to measurements and instructions in a detailed process

No Comments

No Known Activations