INDEX

Explanations

initials or abbreviations that include 'OD' with increasing activation values

oai_token-act-pair · gpt-3.5-turbo

references to the abbreviation "OD" related to 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

Pry

-0.74

 Welsh

-0.71

 Athena

-0.67

 Percy

-0.67

 Pens

-0.65

Ki

-0.65

 Tate

-0.64

 Lerner

-0.64

 Schne

-0.64

 Weston

-0.64

POSITIVE LOGITS

IUM

1.13

OME

1.05

ependent

1.02

ODUCT

0.98

CAST

0.95

MAP

0.94

OD

0.92

irect

0.92

DEN

0.92

gins

0.90

Activations Density 0.014%

initials or abbreviations that include 'OD' with increasing activation values

references to the abbreviation "OD" related to various contexts

No Comments

No Known Activations