INDEX

Explanations

mentions of actions, statements, or decisions made by a specific person

oai_token-act-pair · gpt-3.5-turbo

references to a specific speaker or subject's statements

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

+/-

-0.61

âĺħâĺħ

-0.61

 Starg

-0.59

[|

-0.59

 guiActiveUnfocused

-0.58

 âĢ¦"

-0.56

 ..."

-0.55

 Chaser

-0.55

~~~~

-0.54

___

-0.53

POSITIVE LOGITS

zbollah

1.39

pherd

1.32

resy

1.19

ldon

1.12

ffield

1.11

odore

1.09

ppard

1.06

miah

1.04

idi

1.00

itage

0.99

Activations Density 0.160%

mentions of actions, statements, or decisions made by a specific person

references to a specific speaker or subject's statements

No Comments

No Known Activations