INDEX

Explanations

elements related to legal and criminal activities

oai_token-act-pair · gpt-3.5-turbo

terms associated with significant events or conditions

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

ãĥĥ

-0.70

odox

-0.69

alys

-0.66

ãĥĥãĥī

-0.65

volent

-0.64

ãĥª

-0.60

eu

-0.60

ovo

-0.59

ãĤ¤

-0.58

erent

-0.58

POSITIVE LOGITS

awaru

0.77

 ---------

0.64

(>

0.64

 consisting

0.62

 Annotations

0.61

(<

0.60

izabeth

0.60

perty

0.59

lished

0.57

ahime

0.57

Activations Density 0.897%

elements related to legal and criminal activities

terms associated with significant events or conditions

No Comments

No Known Activations