INDEX

Explanations

references to numerical values or sections within square brackets

oai_token-act-pair · gpt-3.5-turbo

references to data or results denoted by brackets

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

 redu

-0.79

edIn

-0.74

ramid

-0.68

 factories

-0.68

 seams

-0.66

 Elys

-0.66

 handlers

-0.66

mable

-0.66

 ratios

-0.65

 antioxid

-0.64

POSITIVE LOGITS

?]

1.33

!]

1.21

 edit

1.21

sic

1.19

ËĪ

1.17

Footnote

1.16

],

1.10

emphasis

1.09

actionDate

1.08

].

1.07

Activations Density 0.041%

references to numerical values or sections within square brackets

references to data or results denoted by brackets

No Comments

No Known Activations