INDEX

Explanations

terms related to determined and ongoing efforts to achieve specific goals or ideals

oai_token-act-pair · gpt-3.5-turbo

phrases that mention the concept of "pursuit" 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

annis

-0.67

soType

-0.65

iov

-0.64

inki

-0.63

erva

-0.62

 corros

-0.62

 saline

-0.62

ropolitan

-0.61

jab

-0.60

adier

-0.60

POSITIVE LOGITS

eering

0.92

 pursued

0.83

ttes

0.82

hunt

0.81

 pursuit

0.81

eers

0.79

 pursuits

0.75

 tracker

0.73

 thereof

0.72

Goal

0.71

Activations Density 0.022%

terms related to determined and ongoing efforts to achieve specific goals or ideals

phrases that mention the concept of "pursuit" in various contexts

No Comments

No Known Activations