INDEX

Explanations

references to different fictional or real-world settings

oai_token-act-pair · gpt-3.5-turbo

references to fictional or narrative 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

usterity

-0.86

ilee

-0.86

hammad

-0.83

ible

-0.82

idy

-0.75

assic

-0.74

actus

-0.73

agan

-0.72

obe

-0.71

istry

-0.69

POSITIVE LOGITS

 Spray

0.76

 aside

0.74

 forth

0.71

 showc

0.68

 Gork

0.66

 Setting

0.66

 conducive

0.66

ters

0.65

URN

0.64

tle

0.62

Activations Density 0.020%

references to different fictional or real-world settings

references to fictional or narrative contexts

No Comments

No Known Activations