INDEX

Explanations

words related to the treatment or discussion of psychiatric disorders

oai_token-act-pair · gpt-3.5-turbo

terms related to specific plant types or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FUL

-0.73

 Catalyst

-0.70

 Templar

-0.69

 congr

-0.66

Ou

-0.64

renheit

-0.64

fully

-0.62

LESS

-0.62

 Lauder

-0.62

creen

-0.62

POSITIVE LOGITS

olicy

0.97

edia

0.94

onent

0.93

posium

0.90

ublic

0.89

rint

0.88

rompt

0.88

amph

0.87

olit

0.87

ensions

0.87

Activations Density 0.010%

words related to the treatment or discussion of psychiatric disorders

terms related to specific plant types or classifications

No Comments

No Known Activations