INDEX

Explanations

phrases related to significant or impactful actions or events

oai_token-act-pair · gpt-3.5-turbo

the term "meaningful" 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

 Dogs

-0.80

 Tycoon

-0.80

nery

-0.72

bug

-0.66

resp

-0.66

 Fever

-0.66

jack

-0.64

av

-0.62

peak

-0.62

GPU

-0.62

POSITIVE LOGITS

terday

0.94

issance

0.88

itarian

0.85

ively

0.82

 guiActiveUn

0.80

itized

0.79

soType

0.78

ible

0.76

edIn

0.76

ational

0.75

Activations Density 0.016%

phrases related to significant or impactful actions or events

the term "meaningful" in various contexts

No Comments

No Known Activations