INDEX

Explanations

words related to time durations and future plans

oai_token-act-pair · gpt-3.5-turbo

phrases indicating anticipation or future plans

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

xtap

-0.61

 tended

-0.60

uphem

-0.59

 objected

-0.58

 drew

-0.56

 dividing

-0.56

 instinctively

-0.55

 dehuman

-0.55

 differed

-0.55

 copied

-0.55

POSITIVE LOGITS

 tomorrow

0.92

 hereafter

0.89

 someday

0.88

 hopefully

0.84

Hopefully

0.80

eday

0.79

 anytime

0.77

morrow

0.74

0.73

 ASAP

0.73

Activations Density 1.601%

words related to time durations and future plans

phrases indicating anticipation or future plans

No Comments

No Known Activations