INDEX

Explanations

phrases related to future events or expectations

oai_token-act-pair · gpt-3.5-turbo

verbs indicating future actions or projections

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

pmwiki

-0.85

 supposed

-0.75

 supposedly

-0.70

 admittedly

-0.62

 pretending

-0.62

Avg

-0.62

 mater

-0.61

mas

-0.61

 dare

-0.60

 forgetting

-0.60

POSITIVE LOGITS

 tomorrow

0.99

ezvous

0.84

 someday

0.80

morrow

0.77

uate

0.77

stanbul

0.75

igate

0.73

lav

0.73

omething

0.72

dule

0.70

Activations Density 0.391%

phrases related to future events or expectations

verbs indicating future actions or projections

No Comments

No Known Activations