INDEX

Explanations

phrases related to specific tasks or jobs

oai_token-act-pair · gpt-3.5-turbo

references to tasks or assignments, particularly in a structured or formal context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Reincarn

-0.66

 trademarks

-0.63

Geo

-0.61

 chloride

-0.61

orkshire

-0.60

IGH

-0.60

ental

-0.59

 Norn

-0.58

Strong

-0.58

 Neph

-0.58

POSITIVE LOGITS

task

1.15

 task

1.06

masters

0.90

Task

0.89

mire

0.89

 tasks

0.88

lessly

0.87

 Task

0.84

ioned

0.82

nob

0.82

Activations Density 0.016%

phrases related to specific tasks or jobs

references to tasks or assignments, particularly in a structured or formal context

No Comments

No Known Activations