INDEX

Explanations

phrases related to stressful or dangerous situations

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "nightmares" in various contexts

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

ribut

-0.84

gemony

-0.82

rylic

-0.80

ateur

-0.79

eln

-0.77

ribute

-0.77

Reference

-0.76

Ģ

-0.75

bis

-0.75

ensibly

-0.75

POSITIVE LOGITS

 nightmare

1.11

 nightmares

0.89

 scenario

0.87

 Nightmare

0.85

 scenarios

0.79

 Incarn

0.77

mares

0.71

 clut

0.68

 dreamed

0.66

 havoc

0.66

Activations Density 0.021%

phrases related to stressful or dangerous situations

references to the concept of "nightmares" in various contexts

No Comments

No Known Activations