INDEX

Explanations

descriptions of hypothetical or speculative scenarios that involve human actions

oai_token-act-pair · gpt-3.5-turbo

expressions and discussions related to uncertainty or speculation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 respectively

-0.82

}.

-0.66

çĶ°

-0.59

.).

-0.56

%).

-0.52

é¾įå

-0.51

ãĤ©

-0.51

ãĥĩãĤ£

-0.51

çĶŁ

-0.51

).

-0.51

POSITIVE LOGITS

 explanations

0.55

 clearer

0.49

 redund

0.47

 seiz

0.47

awa

0.46

 specifics

0.43

emort

0.43

 proactive

0.43

clusively

0.43

 positives

0.43

Activations Density 4.536%

descriptions of hypothetical or speculative scenarios that involve human actions

expressions and discussions related to uncertainty or speculation

No Comments

No Known Activations