INDEX

Explanations

quotes and statements attributed to people

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or hesitation in speech

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 clone

-0.69

 orchestr

-0.68

 estranged

-0.67

 Reincarnated

-0.67

 endeav

-0.66

 actively

-0.66

 ideal

-0.65

 genius

-0.65

 fused

-0.64

 centrally

-0.64

POSITIVE LOGITS

Asked

1.25

ONSORED

1.21

Newsletter

1.18

Comment

1.13

Write

1.07

Said

1.06

======

1.06

Refer

1.05

----------

1.05

-----------

1.04

Activations Density 0.363%

quotes and statements attributed to people

expressions of uncertainty or hesitation in speech

No Comments

No Known Activations