INDEX

Explanations

adverbs indicating extreme intensity or negativity

oai_token-act-pair · gpt-3.5-turbo

intensifiers that convey negative emotions or descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

iture

-0.84

ioch

-0.72

tan

-0.72

gat

-0.72

oak

-0.71

winner

-0.68

rule

-0.68

claim

-0.68

 Turks

-0.66

olan

-0.65

POSITIVE LOGITS

 terribly

0.97

TPPStreamerBot

0.94

 horribly

0.88

 earthqu

0.80

 badly

0.77

 sorry

0.76

ãĤ´ãĥ³

0.75

 MUCH

0.74

vier

0.71

 cumbers

0.71

Activations Density 0.005%

adverbs indicating extreme intensity or negativity

intensifiers that convey negative emotions or descriptions

No Comments

No Known Activations