INDEX

Explanations

phrases indicating anticipation or expectation of future events

oai_token-act-pair · gpt-3.5-turbo

the phrase "see" and its variations indicating future expectations or observations

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

 contrace

-0.75

 Adin

-0.66

purpose

-0.65

 cons

-0.63

phia

-0.63

 LIFE

-0.61

 tranquil

-0.60

 whistle

-0.59

vernment

-0.59

 task

-0.59

POSITIVE LOGITS

inelli

0.89

 parallels

0.88

dust

0.82

 similarities

0.81

awed

0.75

peak

0.74

agame

0.72

 glimps

0.71

 firsthand

0.71

 signs

0.70

Activations Density 0.075%

phrases indicating anticipation or expectation of future events

the phrase "see" and its variations indicating future expectations or observations

No Comments

No Known Activations