INDEX

Explanations

words related to taking action or responsibility

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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

selage

-0.78

ecause

-0.69

rontal

-0.66

orsche

-0.63

pport

-0.62

ores

-0.60

raid

-0.60

è¦ļéĨĴ

-0.60

 Reasons

-0.59

herent

-0.58

POSITIVE LOGITS

 forth

0.99

frog

0.97

 aside

0.97

 forward

0.91

 ashore

0.87

up

0.84

 foot

0.84

up

0.80

out

0.79

toe

0.77

Activations Density 0.029%

words related to taking action or responsibility

instances of the word "step" and its variations

No Comments

No Known Activations