INDEX

Explanations

broader or more detailed descriptions or comparisons

oai_token-act-pair · gpt-3.5-turbo

adverbs that indicate precision, specificity, or frequency

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

RAW

-0.72

Kh

-0.67

gio

-0.67

Hong

-0.66

Kh

-0.61

 Braves

-0.61

USD

-0.60

 Bomber

-0.60

 Bengal

-0.60

 Million

-0.60

POSITIVE LOGITS

 than

1.29

than

1.06

models

0.75

 oriented

0.74

addon

0.73

apist

0.73

FIG

0.71

 formulated

0.71

akable

0.71

developed

0.71

Activations Density 0.040%

broader or more detailed descriptions or comparisons

adverbs that indicate precision, specificity, or frequency

No Comments

No Known Activations