INDEX

Explanations

references to future plans or potential scenarios

oai_token-act-pair · gpt-3.5-turbo

terms related to planning and forecasting

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

idy

-0.80

idious

-0.79

girl

-0.73

killer

-0.72

aceae

-0.69

ivari

-0.68

 Alph

-0.68

unin

-0.67

lain

-0.66

lyss

-0.65

POSITIVE LOGITS

hower

0.77

urations

0.75

pty

0.75

Ń·

0.74

ãĤº

0.73

ãĤ¤ãĥĪ

0.70

ãĥĥãĤ¯

0.70

ational

0.70

æĪ¦

0.69

andum

0.68

Activations Density 0.068%

references to future plans or potential scenarios

terms related to planning and forecasting

No Comments

No Known Activations