INDEX

Explanations

phrases or sentences starting with specific trigger words

oai_token-act-pair · gpt-3.5-turbo

the word "starting" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

itsch

-0.89

ilit

-0.77

ullah

-0.71

abilities

-0.71

eely

-0.71

abyte

-0.70

Report

-0.69

icular

-0.68

vertisement

-0.68

ega

-0.68

POSITIVE LOGITS

 salaries

0.78

 anew

0.76

starting

0.71

XI

0.67

 point

0.65

ãĤ¤ãĥĪ

0.65

 fires

0.62

 pains

0.62

ģ«

0.61

points

0.61

Activations Density 0.027%

phrases or sentences starting with specific trigger words

the word "starting" in various contexts

No Comments

No Known Activations