INDEX

Explanations

comparisons of actions or levels of achievement

oai_token-act-pair · gpt-3.5-turbo

comparisons regarding the impact of actions or influences on various subjects

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Users

-0.69

Sah

-0.68

icas

-0.64

held

-0.64

 Fever

-0.63

Height

-0.62

urated

-0.62

Parameters

-0.62

resses

-0.62

icia

-0.61

POSITIVE LOGITS

 damage

1.09

 harm

1.08

 homework

1.05

 chores

1.04

 groundwork

0.90

 work

0.89

 mischief

0.88

research

0.88

 reconnaissance

0.87

 outreach

0.85

Activations Density 0.104%

comparisons of actions or levels of achievement

comparisons regarding the impact of actions or influences on various subjects

No Comments

No Known Activations