INDEX

Explanations

phrases indicating some level of certainty or comparison, often involving the phrase "at least."

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a minimum or a baseline condition

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

rence

-0.74

 FANT

-0.70

bath

-0.69

rend

-0.67

icides

-0.65

iard

-0.64

ãĥ¼ãĤ¯

-0.63

bern

-0.63

ses

-0.61

shr

-0.60

POSITIVE LOGITS

 partly

0.74

uner

0.71

 partially

0.69

fair

0.69

 judging

0.69

ety

0.67

 toler

0.65

 theoretically

0.65

 temporarily

0.63

een

0.62

Activations Density 0.023%

phrases indicating some level of certainty or comparison, often involving the phrase "at least."

phrases indicating a minimum or a baseline condition

No Comments

No Known Activations