INDEX

Explanations

verbs in the past tense that describe looking or observing

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "look at."

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

trap

-0.70

Guest

-0.69

Skill

-0.68

pite

-0.65

stress

-0.65

theless

-0.65

widget

-0.64

weather

-0.64

NL

-0.62

hop

-0.60

POSITIVE LOGITS

ĸļ

0.75

ibilities

0.75

=-=-=-=-

0.74

adata

0.70

hetically

0.70

umbn

0.68

 least

0.68

rical

0.68

isions

0.66

iency

0.66

Activations Density 0.046%

verbs in the past tense that describe looking or observing

instances of the phrase "look at."

No Comments

No Known Activations