INDEX

Explanations

instances of the word "somehow" in the text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "somehow" indicating uncertainty or a lack of clear explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

agents

-0.76

oft

-0.71

holder

-0.69

lag

-0.66

Topics

-0.66

pu

-0.65

 stances

-0.65

yer

-0.64

helm

-0.63

players

-0.63

POSITIVE LOGITS

 somew

1.01

 magically

0.90

 somehow

0.90

 mirac

0.88

 managed

0.83

ãĤ´ãĥ³

0.83

 overpower

0.77

Else

0.77

 unaccount

0.75

arently

0.73

Activations Density 0.006%

instances of the word "somehow" in the text

instances of the word "somehow" indicating uncertainty or a lack of clear explanation

No Comments

No Known Activations