INDEX

Explanations

timestamps

oai_token-act-pair · gpt-3.5-turbo

the word "at" 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

FTWARE

-0.74

reditary

-0.72

chuk

-0.67

aceutical

-0.64

selves

-0.64

legate

-0.60

 Vaugh

-0.58

ovych

-0.57

bite

-0.57

 Lucia

-0.57

POSITIVE LOGITS

 least

1.10

mosp

0.97

onement

0.95

roph

0.88

 halftime

0.88

yp

0.85

hens

0.81

letico

0.75

rial

0.75

las

0.75

Activations Density 0.110%

timestamps

the word "at" in various contexts

No Comments

No Known Activations