INDEX

Explanations

phrases related to entering or assuming a more responsible or prominent role

oai_token-act-pair · gpt-3.5-turbo

instances of the word "step" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

eer

-0.66

orsche

-0.66

è¦ļéĨĴ

-0.62

iting

-0.61

udeb

-0.61

 Racial

-0.61

-0.60

oc

-0.60

raid

-0.59

 Rust

-0.58

POSITIVE LOGITS

 aside

1.23

 forward

1.13

 foot

1.06

up

0.97

 forth

0.93

 away

0.91

frog

0.90

 ashore

0.87

 sideways

0.86

 down

0.86

Activations Density 0.021%

phrases related to entering or assuming a more responsible or prominent role

instances of the word "step" and its variations

No Comments

No Known Activations