INDEX

Explanations

phrases related to a series of events or instances

oai_token-act-pair · gpt-3.5-turbo

phrases and words related to reports and statistics, particularly focusing on sequences of events or incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

anium

-0.70

Ability

-0.66

iner

-0.62

66666666

-0.62

 EVEN

-0.61

puters

-0.61

 honestly

-0.61

orge

-0.60

uscript

-0.59

acebook

-0.59

POSITIVE LOGITS

 trend

0.86

 troubling

0.74

 lately

0.73

jer

0.72

 consolidation

0.70

 bolst

0.70

 widening

0.70

 saga

0.70

 controversies

0.69

 ongoing

0.69

Activations Density 0.245%

phrases related to a series of events or instances

phrases and words related to reports and statistics, particularly focusing on sequences of events or incidents

No Comments

No Known Activations