INDEX

Explanations

words related to actions or activities that are being done

oai_token-act-pair · gpt-3.5-turbo

the recurring use of the word "doing" in various contexts

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

lights

-0.75

wit

-0.71

sg

-0.71

RAFT

-0.65

ules

-0.65

Tier

-0.65

pter

-0.65

 Secondly

-0.64

isu

-0.64

liner

-0.64

POSITIVE LOGITS

 brisk

0.83

 alright

0.81

 laundry

0.80

 nothing

0.78

 chores

0.78

 homework

0.77

 something

0.76

 wonders

0.76

 renovations

0.75

 cardio

0.75

Activations Density 0.049%

words related to actions or activities that are being done

the recurring use of the word "doing" in various contexts

No Comments

No Known Activations